RETAIL FOOD PRICE INDEX BASED ON M. S. U. CONSUMER PANEL Thesis ‘or “It Daqm of DB. D. MICHIGAN STATE UNWERSITY Ham Fu Wang 1960 THESlS 7‘3} T-‘x ‘ arms 6 1* M,‘ gr? a n 1 ~51 Mm kzv.ét.fillfiliy ta ' 'I‘E‘é xiii a cut .-- 3,. This is to certify that the thesis entitled RETAIL FOOD PRICE DTDFX BASFD ON' M.S.U. CONSUMER PANEL presented by Hsin Fu Wang has been accepted towards fulfillment of the requirements for Ph.D. degree in Agricultural Economics 4 Ln. _—-*_..4.- ' s. ‘0‘. M L :b RETAIL FOOD PRICE INDEX BASED ON M. S. U. CONSUMER PANEL By HSIN FU WANG A THESIS Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics March, 1960 ‘ C v. _ c i w .3 .nx. .3 bk e t. t E a... m x 6» e ~L. .& .4. .\. ,3. .L w... ma. ‘4'». .U- OU ii ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to all those who helped with the completion of this study and preparation of the manuscript. The author is particularly indebted to his major professor, Dr. Gerald G. Quackenbush, for his careful guidance and constant encouragement during the course of graduate work. Special thanks are also expressed to the other members of the guidance committee, Dr. Raleigh Barlowe, Dr. Harry Brainard, Dr. Thomas Mayer, and Dr. Andrew Brimmer, for their helpful guidance and friendly in- spiration in directing the author‘s graduate program. Thanks are also due to Professor Lester V. Manderscheid for his helpful suggestions on statistical problems. Financial assistance provided by Dr. L. L. Boger, head of the Agricultural Economics Department, made it possible for the author to continue his graduate study. Gratitude is expressed to Miss Marjorie Bacon and Mrs. Mary Par sey for typing the original manuscript and Mrs. Allen Wonch for typing the final manuscript. Finally, the author would like to thank his fellow graduate students of the Agricultural Economics Department who readily pro- vided encouragement, friendship and assistance. The author, of course, accepts full responsibility for any errors which may be present in this manuscript. RETAIL FOOD PRICE INDEX BASED ON M. S. U. CONSUMER PANEL BY HSIN FU WANG AN ABSTRACT Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics March, 1960 C, : , . I.) / Approved # \[KI/‘Z’Lf ('.(II J x'l(:/ t 1 if (I? [(2/ It H (Iii: \‘ '.r ' A.“¥ 5? .. fl ... .i. iL 2321‘... p q ...- , .:.-~ "" -....w-K "‘ .1 V O ‘ - . Q ‘. .3 v. . . PM . 29.. v. ”H. 0 any 5 vs ...“. «\5 O . .v v. ... ...A .1 . PM. 6 .c . r.\ g» u“ .. . e «V .\ .u .pu .\ fl... ... P .5 o . . .Q . .Hhow nu. ‘- Av ¢ e .n‘ ....v ... .:.. ...... ...... ...... ..... b. .1 ... N u“ a \- u‘h .- s . iv ABSTRACT The principal objective of this study was to compile an M. S. U. Consumer Panel retail food price index. The index is a statistical measure of changes in the retail prices of food bought by Lansing urban families. Another objective was to compare the M. S. U. Consumer Panel index with other food price indexes. The basic data for this study were the weekly food purchase records of the M. S. U. Consumer Panel. This panel was not specifically designed to collect data for the purpose of constructing a food price index. The price series derived from the panel data were weighted averaged prices for food items of varying sizes and qualities. Panel prices for selected products were compared with Prices for similar products which were collected from the retail food Stores by the Bureau of Labor Statistics for constructing their U. S. City average index, and their index for Detroit. In general, cereal and bakery products prices were not highly correlated, but the COI‘I'elation coefficients for meat items were relatively high. For most products the mean value of B. L. S. prices exceeded the panel prices. Comparisons were made between the panel prices for Lansing and the B. L. S. prices for the U. S. city average. In order to Compare both levels and changes between the B. L. S. price series for the U. S. city average and the panel series for Lansing, the ' .... f -- *v\\v, ;:si‘..’..:.\ “:5 (...-p I "' " ‘V‘ .‘F ' V ‘ 'v> ‘- " . IL:~ v... be. I‘Ot K. “ . e, . - . ,1 . -. , ... - - . -.:.|....»...€ a mCJLS . . ' ‘ «...-n ha‘vu . Q . —‘;:: -/ -‘~€c: -a|tA :54 , O . g - .. \A 35 :L A:- I;"":--av° ‘ " ' m-‘ui..\...-s “...: tor-c. 's - s.- . . . {.:. o-~:.. \- :~.e. sci..A¢¢ES- When the '. Q n I‘. u. p“ ‘ “‘.e“.- fl!“ . . thyn K.' 6‘... v ... u .‘-. 's “1“ ‘\ Th... A : eXVeD. ’7‘ . .. n ..e ‘ O. Km Cv.‘.D;\-. .. O “ O: 2-- .\‘ I . It»: (:9: ~E‘ , E b aethveE ate: .sc\.e sev-es :h ~13 ..., x“, kd‘ .. ‘> \ .. ‘.I f‘..;i ‘ c 36‘" .Wee‘ \ 41?}. ~¢.e v~ v ‘8» . .i .' ‘~ ‘5” . . ”VJ? s‘ "first index" was compiled. This index was compiled using the same weight and base as the B. L. S. used, but using the M. S. U. Consumer Panel prices in the given periods. The resulting index and the B. L. S. index provide a means of comparison of the price levels and price changes between the panel price series and the B. L. S. series. It was found that the panel prices had similar patterns of movements with the B. L. S. prices for the U. S. city average. The B. L. S. quoted prices in general were higher than those actually paid by the panel families. .When the indexes of each group of products were used, the correlation coefficients of the two indexes were highly significant. The regression coefficients indicated that the B. L. S. prices of each group of products fluctuated more widely than those in the panel series, except the ”fruits and vegetables" group. Comparisons also were made between the panel prices for Lansing and the B. L. S. prices for Detroit. In order to compare price changes between the B. L. S. price series for Detroit and the panel price series for Lansing, a ”second index" was compiled. This index was calculated using the B.L. S. "relative importance value weights" for Detroit, the panel average price (1955-57) as base, and the panel prices in the given periods. It was found that the changes in Prices between the panel for Lansing and the B. L. S. for Detroit were about the same as the panel for Lansing and the B. L. S. for the U. 5. City average. However, the correlation coefficients were higher - - .'._ '_' VP :ar.e'. antes n: . .:......',;,. 341‘ .t . St: 2 Ir. n.4‘nt ., .-h ksav ' 5-: v.‘ s .h V lot—3 lanai. . s 0A y. A ~v ~...:.:~.s nu c pA-te ou h. u . . . .-v~-o; Olga- -v~. v-r'~o H. ""s guy - p) O 0“.- . T‘“ v'. . ' t * ; .. \ .:E . o. _ A" .v-... . A u 4“ ‘(E A ~ - ...; ... lg‘ _ s ‘u a“f‘ ‘ u. ‘- ":fi ,‘ “- ~. “‘ 1““ '1». ..3. ‘ _e O. 4’ ‘ . 0‘, ... p -N‘ :‘A r A‘A‘ Ur .Pe . wrs ' ‘VU.‘ vs t. \ .- .s 5. :_F. ' “‘- T c — ‘ ‘- A. ‘ ‘ Ci‘ ‘rc‘k‘ Q . Li‘s. 53%.“: ‘l “we in, C‘._'e> ‘1. -.‘ 31., .J. . ‘22:- ‘Le ~*.'.L ‘0‘ O” ..‘ .‘4. .‘_‘: 1‘“. .g.‘ vi than those between the panel series and the B. L. S. series for the U.S. city average. These results were due primarily to the fact that the panel prices moved more closely with the B. L. S. prices for Detroit than with those for the U. S. city average. The percentage distribution of food expenditures of the 3-year (1955—57) annual average formed the "value weights" of the panel retail food price index. The actual distribution of expenditures (1955-57) indicated that the panel families spent 30. 7 percent of their total food expenditures on the "meat, poultry and fish" group; 20. 3 percent on the "other food at home” group; 20. 0 percent on the ”dairy products" group; 16. 9 percent on the "fruits and vegetables" group; and 12. 0 percent on the "cereal and bakery products" group. The panel family expenditure distribution was compared with the expenditure patterns of foods for other samples of families. The other samples were the Bureau of Labor Statistics for the U. S. city average in 1952, the National Industrial Conference Board for Lansing in 1953, and the Agricultural Marketing Service for the United States in 1955. The results showed that the expenditure distribution of each major food group for the panel was similar to those for the B. L. S. and N.‘I. C. B. samples of families, but differed somewhat from the A. M. S. sample of farm families. After the panel food price index was calculated, comparisons were made with other indexes. The other indexes were the B. L. S. food Price index for Detroit and the N. I. C. B. food price index for Lansing. ' O ‘. ... ._,, b ‘ 1‘ 54¢-4.-o 1‘. a. . . ' ‘. '1-~o.' .._~_ ‘ ‘ .321. ..C.‘C..c - ‘ b ... .....o‘ q. .Fe V .. > . ‘V s&.. a. .“ :J ~ .' ~ . . ‘ ‘ -. _' I..- W - .\' ‘ “ —\,‘ ‘II-A‘. U‘ v ' l .-. A " rt .3 ....__.. V W I‘“. ' be. \~“-P :7“ ""ex. tr . - _.. . ’ ‘ .. r so ._ ~~ ..E 5.3.: U. .--“ . . ‘r-.-- . ""~===s 3' “re a; . .. Va . . v~._. ‘ ~ v "“t’ 3'8 P‘s. - _ e ‘.\'h: 1‘ q .... ‘ v “ ‘ ~ ’ . ~-t ..IHE' “z-” c og—‘_"‘ ‘ 0 h ‘ .'A . ~ . ".‘ ts :Ff‘ h P‘M ~‘ A iv. These comparisons indicated that the panel food price index moved parallel to the B. L. S. and N. I. C. B. food price indexes. However, both the B. L. S. and N. I. C. B. indexes fluctuated with larger amplitude, but less frequently, than the panel food price index. Presumably the greater frequency of variations in the panel index reflected more of the effects of the prices of ”specials" and, to some extent, changes in the quality of some of the items purchased by the panel families from one period to another. In comparing the panel food price index with the B. L. S. food price index, the panel index provides a more accurate measure of the costs of food since the spending patterns represent the actual purchases of the panel families during the 3-year period 1955-57 and prices are those actually paid by the panel families from week to week. On the other hand, the panel price changes were affected by changes in qualities and quantities of food items purchased by the panel families from week to week. .FJ 3' FROCED‘ O-o‘ O (-9 (ll J. (,1) (“U .H ,1. ‘7 viii TABLE OF CONTENTS. Chapter Page I. INTRODUCTION ............... . ...... . ........... 1 Consumer Price Index and Retail Food Price Index . 1 Purposes of the Study .......................... 1 The B. L. 5., N. I. c. B. and A. M. s. Indexes ....... 3 A Critical Review of the B. L. 5. Index ....... .. . . . 6 Sources and Preparation of Data ........ . . . . . . . . . 9 The B. L. S. "relative importance of value weights" 9 in the retail food price index for Detroit The value weights and price base of the B. L. S. retail food price index for the .U. 5. city average 10 The M. S. U. Consumer Panel data 10 II. PROCEDURE AND METHODS ...................... l3 Adjusting the Data .............................. 13 Compilation of the Indexes ....................... 14 The first index. --The index which was compiled using the B. L. S. weights and base (in the retail food price index for the U. S. city average), and the M. S. U. Consumer Panel prices in the given periods , 15 Selecting food items Deriving the B. L. S. price base Deriving the B. L. S. value weights The index formula The linking procedure The seasonal food O‘U‘IrhboNu—o The second index. --The index was compiled using the B. L. S. "relative importance of value weights" (in the retail food price index for Detroit) the M. S. U. The 1': '§ —.’ l’t: ,.~ u‘.. [u Chapter Consumer Panel average 1955-57 price as base, and the M. S. U. Consumer Panel prices in the given periods ’ l. The linking procedure 2. Conversion of base 3. Twelve months, compared with 13 periods The third index. --The M. S'. U. Consumer Panel retail food price index I. Weighting and selecting food items The formula used in calculating the M. S. U. Consumer Panel retail food price index 3. Prices of the items used in the panel index calculation 4. Calculation of the index 5. Selection of base 6. Seasonal food 7. Relative importance. of items III. . COMPARISONS BETWEEN THE M. S. U. CONSUMER ‘ PANEL SERIES AND THE B. L. S. SERIES FOR. U. S. . CITY AVERAGE ................................... The Levels and Changes of the Two Price Series The Relationships between the Two Price Series ..... The correlation and regression coefficients be- tween the two series using original prices The correlation and regression coefficients between the two series using their indexes IV. COMPARISON BETWEEN THETM. S. U. CONSUMER PANEL‘SERIES AND THE B. L. S. SERIES FOR DETROIT The Changes in Prices between the Two Series ....... The Relationships between the Two Price Series. ..... The correlation and regression coefficients be- tween the two series using original prices ix Page 2.1 23. 37 38 51 ’51 57 64 64 75 75 I 32t- .- : New. q-ov 9‘ an—\ .:E Lkra. hv- (H y\ . ‘ :.. .:.R r». mom's; H :w :‘m I~.~ . 1... ts at: A UK- $1.4.“ ..c . V t ‘. A O‘.‘ Chapter The correlation and regression coefficients be- tween the two series using their indexes V. THE EXPENDITURE PATTERNS OF THE M. S. U. CON- SUMER PANEL FAMILIES COMPARED WITH THE Exu PENDITURE PATTERNS OF OTHER SAMPLE FAMILIES How the Panel Families Distributed Their Expendi- tures among the Various Food Items . . . . . . . . . . . . . . . . The Actual Distribution of the 1955-57 Expenditures amongtheFoodItems ..... ........ ..... Changes in Relative Importance over Time . . . . . . . . . . Comparisons of the Expenditure Patterns between the Sample Families of M. S. U. Consumer Panel, B. L. S. Nolgch.,andAoM.S. .....OIOOOOOOOOO...0.0.0.00I VI. THE M. s. U. CONSUMER PANEL RETAIL FOOD PRICE INDEX 0 O O O O O O The Usefulness of the M. S. U. Consumer Panel Retail FOOdPriceIndeXOOOOOOOOOOOOOOOOOO ..... 0.00.00... The Limitations of the Index . Limitations of measurement Limitations in using the index Analyses of Changes in Prices of Food Groups ....... All food at home Meats, poultry and fish Dairy products Fruits and vegetables Other food at home Cereal and bakery products Page 81 86 86 86 88 91 107 107 108 108 108 109 109 111 113 120 I25 126 Chapter VII. VIII. IX. COMPARISON-S OF THE INDEXES CALCULATED USING DIFFERENT PRICES AND WEIGHTS, AND THE SAME PRICES BUT DIFFERENT WEIGHTS, FOR ALL FOODS ..................... . ........... Introduction ................................... Comparisons Of the M. S.RU. Consumer Panel Food Price Index with'the B. L. S. Index for Detroit, and N. I. C. B. Index for Lansing ...................... Comparison of the M. S. U. Consumer Panel retail food price index and the Detroit B. L. S. retail fOOd price index Comparison Of the M. S. U. Consumer Panel retail food price index and N. I. C. B. retail food price index, for Lansing Comparison of the Indexes, Based on the Same Prices but Calculated by Using Different Weights ......... THE STRENGTH AND WEAKNESSES OF THE M. S. U. CONSUMER PANEL RETAIL FOOD PRICE INDEX IN COMPARISON WITH B. L. S. RETAIL FOOD PRICE INDEX A ............ .H.. ........................ The methods of Obtaining the spending patterns from a representative sample of families The techniques of choosing the items which are priced in the index The method of collecting prices The index formulas used Weights used in the index construction SUMMARY AND CONCLUSIONS ..................... BIBLIOGRAPHY . . . . . . . . ................................... APPENDIX ...... . .............................. ' .......... xi Page 148 148 148 148 149 153 160 162 164 164 166 I 167 170 193 197 -vy “ ....- >l- ‘.. mw-‘v- - "9 '3- ...m..:.\ .2. .t_ an: :a-terx' p: \‘mm‘. . \IQ ... V‘JAIA013.\ . ‘ ‘M ‘.. . ' CNNDL.‘ V‘ . e r _ . V"‘ J'. t‘: :5 3. L. S. Elrlk cé ‘V IV." 'V‘v '- o u‘......1‘ . ‘,. V L 1 .L,. “’-,- ‘..-, .. :JAJCA\,LS‘ "E I ‘F.fla' _"“.‘ . 1..-:11 C1.L...3't':'ll 3‘1" 5, U. COWCH“~ ..- “..| Q \-~..,. _ ‘ o “d ‘n -o IOOA“‘r\ v‘ I. ."D n .. i. . 1:” H;"w . And-.:.c“ Dr . e ..A n y. C I .:.:g B L E ’ ‘ Dz.- . ' . A...;:: C." c .‘n ' JO oaae B L g P s . V. 51 i “W - undumqrsv P; v . J. -‘ 5- ' ~‘-\‘ _ I . , E J‘Hs 0‘ “v- <. ' . A 5" r‘ e‘fl‘ . a .. 4 t. .EX C; ‘ . ‘QCI‘ _“o'.‘ n ‘ ‘.. ¢"_ Taaae r. J; " W .Ln A-E,HS hf" I Fr, . J‘c I .\.. ‘ a .. ~‘ '2‘. .t.“ F V- ”C‘s; ‘11“, s. "E'..‘fi Q. \t: Table 10. xii LIST OF TABLES Page Summary of regression and correlation results for cereal and bakery products, using B. L. S. prices for U. 5. city average and M. S. U. Consumer Panel prices, 1955 to 1958 . . 53 Summary of regression and correlation results for meats, using B. L. S. prices for U. 5. city average and M. S. U. Consumer Panel prices, 1951 to 1958. . . . ................. 54 Comparison between the me an of Panel prices and the mean of B. L. S. prices for U. S. city averages .................. 54 Summary Of regression and correlation results for groups of products, using the index for U. 8. city average and the index calculated with the B. L. S. weights and base, and M. S. U. Consumer Panel prices in the given periods . ...... 58 Summary of regression and correlation results for cereal and bakery products using the B. L. S. prices for Detroit and M. S. U. Consumer Panel prices, 1955 to 1958 .......... 79 Summary of regression and correlation results for meats, using B. L. S. prices for Detroit and M. S. U. Consumer Panel prices, 1955 to 1958 .............................. 79 Comparison between the mean of Panel prices and the mean of the B. L. S. prices for Detroit .......... . . . . . ....... . . . . 80 Summary of regression and correlation results for the groups of products, using the B. L. S. index for Detroit and index calculated with B. L. S. relative importance of value weights for Detroit, and M. S. U. Consumer Panel prices in the given periods and 1955-57 average prices as base ....... 80 Table of items priced for the retail food price indexes and their relative importance to total food index, selected dates, M. S. U. Consumer Panel, Bureau of Labor Statistics, National Industrial Conference Board, and Agricultural Marketing Service ...................................... 98 M. S. U. Consumer Panel retail food price index-agroup and subgroup indexes of selected foods. ...................... . 138 a u... . o P.~';..5A ‘ bdlctvfi‘. u-v ..- r- o... :4.» 0.“ :4“. . “...; ..- n y . .:cho-V \‘ , ham-e...“ . n.‘... ~. C;. v .- :.v‘oa- Al . O . O > 0.. ,. ”'y O "' '\ ... a n 1' ' ""‘ "r . ..e‘\.i- . ‘- m... ..., s . .. C""“:.>~‘ ' . us..v“.--b- . -- .I I A L - O -945 _a “r a u l S“‘v JD... r'~.r v.1, -p-""~ . t-I \ I ut.-_‘. .. ..I. x a 9 . d.~'_':‘ " :Jv . 0 o ..._._‘ ’ I I T . ‘ '~.~~ . ~‘.‘=.“ \ . ' ‘9 - ..\\_. V . -...u;:‘c , . s 4‘. a- p... . "on “"He A..H__ J:.,',t—f.v- 5‘. \. - ‘ I " ”um?”- ~ ‘ sn" ,4 n \— I an. ‘ C-Mu- a.‘ .41.. . ‘sac —. - .... I -- HV- 4- ' - t.‘ cc '0 . " .1 ‘ *n . ..v‘ ‘l! ‘ ‘4. L:- ' ‘.‘S w~ ‘ ... . 5. u e. - A w ' 9 . _ . C. Vs. . "-.. ..p. ‘4 ‘ HS, - . a.‘ .‘ a. s". R”; ... . g . r1 0“ .' .. ‘Q: -‘ a ~‘~.:"‘- ‘, a. . .K‘ O\ I '- ~v eg- 'I .. C- Fl- “.Mau. A ‘A‘S‘— ‘ D a. -. a r". - " 1 u‘. 3’9. '. ‘ t.w'. "(.9- f. . ~ - u“ .‘ ‘ "‘¢:;I- t-n. ;-.. ‘5 .-‘ - u. — “-J“Ic l I - ‘Va ‘ o n, so \:~ .. ‘ -, :E"AF Q ~ a . VA > h ~S ._ . .J» . .I‘ w, "‘scecn 3‘4. ,3, ' I! Figure 10. LIST OF FIGURES Comparison of the changes in retail prices of meats, poultry and fish between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city average . Comparison of the changes in retail prices of dairy pro- ducts between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for city average .................. Comparison of the changes in retail prices of fruits and vegetables between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city average ............. Comparison of the changes in retail prices of cereal and bakery products between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city average . . . . Comparison of the changes in retail prices of other food at home between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city average ............. Comparison of the changes in retail prices of all food at home between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city average ............ Comparison of the year average of the changes in retail prices of food groups between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for U. S. city ‘ average 0000.00.00 000000000 0 OOOOOOOOOOOOOOOO O 00000000 Comparison of the changes in retail prices of meats, poultry and fish between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for Detroit, Michigan . Comparison of the changes in retail prices of dairy pro— ducts between M. S. U. . Consumer Panel for Lansing, Michigan, and B. L. S. for Detroit, Michigan ............ Comparison of the changes in retail prices of fruits and vegetables between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for Detroit, Michigan ............. xiii Page 40 41 43 45 47 49 50 66 67 71 n.¢ .\.. -.....v-L .‘5—r‘ - y.-..~.o. ‘ H‘ a - V‘V' .... .,.. u. a‘aosceo L '5 ~..: p I D‘ . . w.- ... ..L C . . T. c. .\ .F» ..u. . q C S: S I» .... lat». n.. e r . .1 .. RC ... ... .’oo .K o . e .s A .A‘ v. D C ..v to ill! Figure 11. 12. l3. 14. 15. 16. 17. 18. 19. 20. 21. 22, xiv Page Comparison of the changes in retail prices of cereal .and bakery products between M. S. U. Consumer Panel for Lansing, Michigan and B. L. S. for Detroit, Michigan . . . .. . 72 Comparison of the changes in retail prices of Other food at home between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for Detroit, Ivlichigan . . . . . . . .. . . . 73 Comparison of the changes in retail prices of all food at. home between M. S. U. Consumer Panel for Lansing, . Michigan, and B. L. S- for Detroit, Michigan .. . . . . . ........ 76 Comparison of the year average of the changes in retail prices of food groups between M. S. U. Consumer Panel for Lansing, Michigan, and B. L. S. for Detroit, Michigan. 77 Changes in the average retail prices of all food at home, M.S.U. Consumer Panel 110 Changes in the average retail prices of food major groups of meats, poultry and fish, and dairy products, M. S. U. Consumer Pane]. 0000000 o o o o u a o o o ooooooooooooo . - u o c o o o o o 115 Changes in the average retail prices of food subgroups of meats, and fish, M. S. U. Consumer Panel . . . . ........ 116 Changes in the average retail prices Of food subgroups ofpoultry, M.S.U. Consumer Panel 118 Changes in the average retail prices of food major groups of fruits and vegetables, M. S. U. Consumer Panel. . . . . . . 12.1 Changes in the average retail prices of food subgroups of fresh fruits and fresh vegetables, M. S. U. Consumer Panel 0 0 0 D 0 h 0 0 o 9 0 0 0 0 C 0 D 0 O 0 C! 0 0 Q 0 ° 0 0 U U 0 L U 0 9 U V v 0 C' O 0 U 0 L 0 1) 122 _ Changes in the retail prices of food subgroups of canned fruits, canned vegetables and dried fruits and vegetables, MOSQUG cons‘xlmeripanelcoco-JOUOCOOOOOCDecouuououooooo 123 Changes in the average retail prices of food subgroups of frozen fruits and frozen vegetables, M. S. U. Consumer Panel 0 0 0 0 0 O O 0 O O 0 0 0 0 0 0 O O 0 C C 9 v 0 O 0 O 0 s) O G 0 D d O U 0 C- OOOOOOO 0 J U 124 ... .‘ a v. o . 0 n .G r” .r‘ . . .h~ v . nu .5. .1 .5. ‘L .w; LL .C .1 uuh~ VA v.~\ I. .n a. 2 at 1.. .L ... .. .nu .. . .-H .u» n u l by .... n. ..a u I a U a. .v- ~5‘ ...-..‘ u.'- \a no V s v..\ ... {a . g A‘- I\¢ Lu ‘- :: ... .3 .1 .... .... s... PM L. K. V Q U . \I ..4 r. . . _ ‘o .s» .... w . u. . a. .5 r . . a s . u.. .1 .n.‘ W. .A . a P. . i .w. . a .. ‘5 M.- ..» Mu ... .2 A. Figure 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Changes in the average retail prices of food major groups of cereal and bakery products, and other food at home, Me So U9 Consumer panel 0 0 9 C 0 G I'-' 0 O O 0 U '0 0 G O 0 0 0 i v D 3 O Q G O 0 Q U Changes in the average retail prices of food subgroups of cereal and bakery products, M. S. U. Consumer Panel .. Changes in the average retail prices of food subgroups of beverages, and sugar and sweets, M. S. U. Consumer P an e 1 0 K. 0 0 u D a C u L o 0 L a L u u u 0 u u x. L‘ (4 a O c a i; (a 6 G C 0 LLLLLLL ' v C ‘a .. U ,_, Changes in the average retail prices of food subgroups of cooking aids, partially prepared food, M. S. U. Consumer panel 0 9 0 0 0 O C: O 0 D U 0 0 0 U U Q Q 0 C u 0 u v u L‘ J U 0 O a J u {a 0 O 0 3 .- U i. {r uuuuuu ‘J 0 Changes in the average retail prices of food subgroups of fats and oils, M. S. U. Consumer Panel . . . . . . . . . . . . . . . Changes in the average retail prices of food subgroups of eggs, M.S.U. Consumer Panel Changes in the annual average retail prices of food groups, Mg So U0 consumer Panel a U 0 0 U 0 0 ~ 0 U 0 'J D 0 0 0 9 O 0 0 0 0 v Q 0 U 0 v Q 9 Changes in the annual average retail prices of food sub- groups, M.S.U. Consumer Panel Comparison of the retail food price indexes, calculated by using different samples, between the B. L. S. for Detroit and the M. S. U. Consumer Panel for Lansing . . . . . . . . . . . Comparison of the retail food price indexes for Lansing, calculated by using different samples, between the National Industrial Conference Board and the M. S. U. consumer Pane]- C O n O 0 0 v 0 0 L‘ 0 O b ‘0 G 0 O 0 O Q 0 C 0 .2 0 C 0 O 0 O C 0 4 o L 'l' d 0 Comparison of the food price indexes, calculated by usmg B. L. S. weights for Detroit and M. S. U... Consumer Panel “'Eights for‘ Lansing 0 O 0 'v v 0 Q 'J 9 U 0 0 0 O O 0 A Q L C v 0 O C (— O 0 g. v C‘ Page 128 129 130 131 132 134 136 137 150 155 ..IE.."‘..E'? DILCE 1' .1;....M;.‘_.;.'.r‘ - ”nun u.-¢-o.“..‘.on 3 | 75; ‘2 :. WL- .. “"‘*S- ..C” rt, . ‘l‘ h .‘ I ~v... '7 V' v ' ~ - C ...e 53‘1‘1; w‘ :5 _" n. ‘ a F vp-- -.. : Ws - “" “Anne :5 1 ” ply CHAPTER I INTRODUCTION Consumer Price Index and Retail Food Price Index Price indexes used in demand analysis are of the type known as consumer price indexes, which serve to measure the changes in the cost for maintaining a certain standard of living. In empirical demand analysis, an important point is the distinction between nominal and real values. When representing demand as a function of prices and income, the variables may be specified either as nominal, i. e. , prices and income as actually observed, or real, i. e. , the nominal values divided by a consumer price index. Among all items, foods are a very important component of the consumer price index because they account for a major part of total family spending and their prices change very frequently. A retail food price index is simply a part of a consumer price index. It is a statistical measure of changes in retail prices of the foods bought by COnSuIners. Every index is specifically designed for a certain measure- ment, Consequently, it must be applied carefully when used for other purPOSQS. Purposes of the Study One of the primary purposes in this study was to compile a reta‘ . . . . 11 f00d price index based on the Michigan State University Consumer , The irx‘lt' 3‘4 I” ' . ‘ 1 .- Fvu‘ A. . It‘s ..., . ”e a, ...- ..A‘.‘ 0 . . v L-u- ;« 4-3.... "K O 9' I ‘0“..\: u.=-..un.e ... _.~" \ a :P' .‘J‘ “Or. h ' ......-..SJ.. “.... 1.x C “.:.;- QM-, .,,‘; ...-.....u...“ e S__C.C S . axes o: ‘*e .\i' h. 4: . _ ....nJ.‘ b:&°‘51;t5 : . .1... \ v ..i.c, 5, t, '7“,- l..& .53: “‘!A . . i- - 1 ' . ‘ ‘ -=*-:ll mac 0: ??Ir .' ..., .:E fitfect ‘ ‘ ' ' 0. 1:115 DE: \n w. “‘. “6 a E“LS 02 Lars. "-1, ..1l;;$ ‘ «EL-1e du‘ev ‘a‘ ‘ ev- . :v‘.:.'~ s _ '~...) fi...’ A .. . v tA‘EIel“ Pfif AA lu‘ \‘ I“) . 5495510“ d?‘ . ‘ 1.1. Pr . “1_\ x, . gtfu\ EEK b ‘F ‘v ‘ ‘.‘.I. u" :‘::€.‘s¢ .,,: Panel data. The index was a statistical measure of changes in the retail prices of foods bought by Lansing urban families. Emphasis was placed upon the methodology of the compilation of the index and upon the analysis of price movements for groups of products. A second purpose of this study was to examine how Lansing families distribute their food expenditures among various items in comparison with the distribution of food expenditures observed in other expenditure studies. The comparison was mainly based on the sample families of the Michigan State University Consumer Panel, the Bureau of Labor Statistics (B. L. S. ), the National Industrial Conference Board (N. I. C. B. ), and the Agricultural Marketing Service (A. M. S. ). The third purpose was to compare the store sample used for the retail food price index prepared by the Bureau of Labor Statistics with the panel sample used for the retail food price index prepared by the Department of Agricultural Economics, Michigan State University. The object of this part of the work was: (1) to compare the price movements of Lansing, Detroit and the United States city average between the B. L. S. store sample and the panel sample; and (2) to evalluate the difference between the two series of indexes as a result 0f using different methods of collecting prices. Graphic examination and regression analysis of the data were used to determine the relation- shlps between the two price series, as well as some of the differences preSent. - . .._, ,.. .- _\ g . "‘" -:-:a M:- F. a ... ,. . . 1 - 7- “m I-‘m- 4‘ ." ‘iv-n. Lay‘s-C “ ¢- .‘ ' 9 0' ' : -;‘-w, .... “‘1 v .»..—a Cu..,t ‘.-.1.:p. ‘..,. a ‘s-‘vs ‘\0 ~.o,‘, w _ ‘ ‘Q‘ ‘.. g. V. A .:. r t I. \. 4‘} b d I ,N. Z: 5., . “,“n h... ’ -~ 1‘ b ‘t \. . . "u: .. ‘; \'~r ‘;... \.I . v. s38 TC . f“ 5 4.. 1 P o .- .. ‘l ‘N .“c The B. L. S. , N. I. C. B. and A. M. S. Indexes The retail food price index, as calculated on a monthly basis by the B. L. S. , is the best known and most widely used statistical device for measuring the price changes of foods from the past period to any later date. A sample of 81 food items, excluding meels eaten away from home, was used in the B. L. S. revised index (1947-49 = 100) to represent the "market basket" of foods comprising the pattern of purchases of city workers' families in 1952. Expenditures for items in the "market basket" were based on an "expenditure survey" of 8, 000 families in 97 cities during 19502. The cities for the sample were selected to represent different kinds of cities, taking account of city characteristics which affect the way families spend their money. Since the expenditure records were for the year 1950, the "market basket" representative of the year 1952 was adjusted for changes between 1950 and 1952. The average size of the families included in the index was estimated to be about 3. 3 persons, and their 1952 average family in- come after taxes was estimated to be about $4, 160. The method of calculating the B. L. S. index was essentially that of a Laspeyres' method. The index was a chain index, each link \ 1Bureau of Labor Statistics, U. S. Department of Labor, WChapter 7, from B.L.S. Bul. No. 1168, Techniques of W Major B. L. S. Statistical Series, pp. 11-12. 2 Ibid., p. 7. 3Ibid., p. 3. .: q n p g: . ‘ r- 3‘! “it: {burl ,' 1 .....i . . . A .....uu-vy v-n Ou- 0\w¢ -.....u ...ng‘u.‘ 5.. AA. . . ; .n..., w .’ ~ K —-‘....‘,_‘E x". P‘ V» . "‘ an. a ‘ A t _ ~- _ . - ~t ...:rhEt 3:5.‘c0 -."‘" V ~-=:.:“'r U C “' ~ue ~.. .U. Lbs “A ‘ . ., 2 v“. _. . . """“:'e {C 9‘» ...e u, c. ...-.‘e .:. New Y p« Ace C-Loé' fl; . "‘A‘-"l~ u razec 09' ‘8 a ' .. ‘5‘. IV ‘1 . I. ‘- ' ' W1“. ‘1» ‘- 4 calculated by the Laspeyres' formula. The food prices were obtained monthly during the first three days of the week containing the fifteenth of the month. They were collected from 2, OOO chain and independent retail food stores in 46 cities. The value weight was the proportionate expenditure in the "market basket" for the subgroup which each item represents in the expenditure survey. For example, the weight assigned to the price of white bread is the proportionate expenditure in the "market basket" for all bread and plain rolls. In combining the cities into the U. S. city average index, each city was given a weight proportionate to the wage-earner and clerical-worker population. For example, the relative population weight of the urban area of New York City was 12. 5 compared with 3. O for Detroit. 4 Therefore, the average price change in New York had about four times as much weight, in the U. S. city average index, as an equal price change in Detroit. Besides the index for the U. S. city average, the B. L. S. retail food index was also separately calculated for the 20 large cities. The index for each city was calculated monthly in terms of its own weight and price change. Detroit was one of the five largest cities in this group of 20. The N. I. C. B. revised retail price index for food at home was composed of 78 items. The value weight of the revised index was derived from the consumer expenditure survey of 1950 made by the x 4 Ibid., pp. 8-9. ELEf The basmc dat. :‘as‘xveyed Citv TLCB. expendztures :e::.t.:e data for ct}; The rex-zseri : .1". {r ...: g; (D m (J 0.1 ’1 (I. D O O I :c:s:::1.a:.sir.g wen IV“;~\‘A V ' ‘ .........e. ARC Decemse ~. .I a [he 'v‘n‘.‘ L Y f¢u~s..8d. in gar-1 ‘u ~n ".‘cex prepared b\' t" ..2": “J'C . ‘ p .m e.r.:ex tor La: The we‘ *1“! 3.51:3" .' , -p.1ces Said 5.. . o ‘ 4 1“ ‘N Elm—M... ““;ure sur‘¥'e‘ ‘::T .3" 6L , } the A. hi. S ' I 4 (“L ‘ME “v- .. , Enid; 0:1}‘8 C \.:1' 3‘ 6' , A B. L. S. 5 The basic data from this survey were adjusted. Since Lansing was not a surveyed city in the B. L. S. expenditure survey of 1950, the N. I. C. B. expenditures for this city were interpolated from existing expenditure data for other cities. The revised index (1953 = 100) for the U. S. city average covered 40 cities. Prices were collected monthly, quarterly, and annually by the Board from a sample of stores in each of 40 cities. Prices for Lansing were collected four times a year: in March, June, . September and December, the months when the food price indexes were published. In general, the N. I. C. B. index was very similar to the index prepared by the B. L. S. Of particular interest was the N. I. C. B. food price index for Lansing, which was not prepared by the B. L. S. The weighting pattern of the recently revised A. M. S. index of prices paid by farmers for food and tobacco was based on the farm expenditure survey which was conducted in the spring of 1956 J"Dintll’ by the A. M. S. , the A. R. S. (Agricultural Research Service), and the Bureau of the Census. The data were supplemented by a survey of food consumption made in 1955 by the A. M. S. and the \ 5National Industrial Conference Board, Development of the Wet Basket (New York: National Industrial Conference Board, P. 1- 2. 6 I National Industrial Conference Board, The Market Basket 01' Unsurv In\\¥eyed Cities (New York: National Industrial Conference Board, " 1955) pp 1- 3 7 Price In National Industrial Conference Board, Revised Consumer 1954) pdex (New York: National Industrial Conference Board, Inc. , P- 304. ..4\ \o C. v .C .»u be .... u v T F 5‘5 .\¢, 5 , a‘. O . .r“ “F” :3 C A Q. .1 e u» a“ e T . 3“ n .... .. 9 .0" .- «. ‘4 4....-. o‘- A ‘3 .. . . in C ... 1. S ... c... .. 3 C v. n\ e h“ 1 . e .1 .6 .... by .... .2 ... I .1 9. .1“ L... . . TA .6“ Hum ...N; .e e .‘u R .s a» on El n.“ L... I . . E a. .. e ... . ..u. ..u ...w. a: a.» L“ :. ... . . . s . .u R» .\ at a . ... \. :- .. u. .s 5.. 5~ . A r . .. ,.~ ... .... .. .. . )x u"... u... ax .2. u . _\ A. R. 8..8 Food was combined with tobacco as one group in the index. The A. M. S. food and tobacco price index was composed of 52 items. The 50 food items priced represented 91 percent of total expenditure weights, and the two tobacco items accounted for 9 percent. The index was specifically designed to measure the food and tobacco prices paid by all farm families. A Critical Review of the B. L. S. , N. I. C. B. and A. M. S. Indexes The merits of the B. L. S. consumer price index have long been a controversial issue. The following (brief discussion from theore- tical and technical points of view touches only some of its weaknesses with respect to the food price index. The changes in weights and the way of collecting prices are connected with most of the problems in the B. L. S. index. For example, the 1952 weights used in the revised indexes were derived from the 1950 consumption patterns. Food expenditures between 1950 and 1952 were adjusted solely for changes in price. This, of course, ignored the fact that the food consumption (for "pricing families" in 1952 was greater than in 1950, both in money and in real terms. 8B. R. Stauber, R. F. Hale and B. S. Peterson, "The January 1959 Revision of the Price Index, " Agricultural Economic Research, Vol. XI, Nos. 2 and 3 (Washington: U. S. Government Printing Office, 1959), p. 35. 9Ibid. , p. 52. 10Willard W. Cochrane and Carolyn Shaw Bell, The Economics of Consumption (New York: McGraw-Hill Book Company, Inc. , 195(3), p. 72. "0 was d-e to fit. . ...-I.\DIA ,. -...,'.Q 0.. . n .7' .-z........ ... .I'...4¢. \ '3'.'L"..'€ «Kai's :r. t :ttencztzes were he. ' u "; \ -. . , ”"“"~ v...\ a\e:3gr ~;'.""‘Fa 6‘ - II “«>..1“~"“ unfit S ..v 'v‘...’ ‘ . ..A ‘.""P---.. .‘ “‘... i -. r ....... .:1..... . m- . D ... “-l 'e‘ ‘0 .o . a... u L. .. '11 t 1 ('1. Ln .“41 ‘ ‘ V‘V‘P. ‘ ‘y a n . “*SEC 3” .. .. I I -‘l .‘--.' .V ‘ "s. has ~ - Lace Szm . g. ‘\ .:...- ...‘,_ v F D "-1,: .. .t'e 3‘ ,1." . in“ ' f ‘& \ “(F-u »‘ ., ...“ . A.d.es “‘QL ‘ 1 “A. v\ ‘c .5. S: 2 ._ . EL\ :‘Fr‘ ’ . x~ e‘.1 c“ ., ~. ‘ ~ ‘1 5: ...“; ...“ ...u. :L ‘Jf‘~‘ .| 'r - . __ 9“ '5 . '-. 3.. . ‘H:FC . ‘. P~.:€.: . - C - . .‘\. ‘H T‘ ‘H ‘ ‘5’ t1‘4: ~ " 1 2‘ CI ‘5 The latter was due to the increase in quantities consumed without taking into account the influence of changing prices. When changing the expenditure weights in the revision, the population weights among different cities were held constant. This could destroy the significance of the U. S. city average index in representing the changes in the cost of maintaining the same level of living. If expenditure weights are changed without taking into account the shifts in population, the index does not reflect its real meaning because the weighted average price of a commodity for the U. S. city average depends not only on the price quotation in each city but also on the weight of each city in terms of its population relative to the nation as a whole. Even if the prices remain constant, the U. S. city average prices do change if the city weights change. In comparing the panel series and the B. L. S. series, the B. L. S. quoted prices are usually for food higher in quality than those actually purchased by the M. S. U. Consumer Panel families. It is doubtful that the sample of food items actually selected is truly representative of all food bought by city wage-earners and clerical- worker families with “moderate income. " As we know, food prices change very frequently. 'The B. L. S. food prices are collected only once a month, about the middle of the month. It is conceivable that the collected prices could fall in an extremely high or low period of the month. Thus, the changes in food price indexes from one month to another may not represent the monthly average price changes. In -‘e 3,1,5. 2; “...- ‘ .' in ‘9 v::“"’f\\€€f( “5"“ ‘ ,.',_.~,:|.9 De mjre 6x ”... ”...-'- L:..::.:E :1. The NW. C. l ;-,o..~ p-OO y'- ‘ W '5 -.;..,...i .3».€L-.S, V - ' n I ;.‘ ‘A- .. 0.. \ . -.. :vu.i u: o . A. n .- , ._.\‘ ' F ' _ ' ' ‘:"IA|L.B. .:.:t‘k ls : .1. .- .._ 4"\_ ‘." “‘1 V‘ e J CAL.CS \ ”:7. ' " ‘v -v v-v- ‘u. x . ....1 ._, . “c‘e u. a. .. '3: no}. .- .1r_“ {r . ...... 4...e. “-3.. ...‘ I ‘1 M‘ 1 e ‘ ~ (H P, Q . . ...C ...‘ee-m ’v- r '. .'§g.- I g . \"u.':: A. 'Fnfi- . Va ‘Vv‘: I C5 -‘t .. v...u(‘.‘ ‘41." b ';'-. ‘ ‘ :- .. . a. .:‘ .I-E 1“.“A . ct “““5 Ca,“ ‘4‘. \ . ... _ \‘ I.“ h... one! EEQE‘;*’»., y u‘ ‘rc.‘l\ .‘ 5. . ". J“W~—-‘ ‘AS‘I‘: ““Ce ‘ ‘o . ‘,. x. ._ p4 ‘5».11 rt ‘ 4‘ \IA -1 42:...- -. . ‘97. ‘J.\ 1.. A ’1 a.“ - . _ ' re [1‘6” ‘ ._, ‘ ‘ 't ...“;i; .‘.., ‘ ‘~ ... I.‘ . ‘ c. we 1r'— if C . : L . ‘,. \‘ “A 'A I l‘:e;.e . ‘ - . 31‘»... “ K a”, -'. .‘ addition, the B. L. S. food prices are obtained during the first three days of the week when few of the food items are bought. The B. L. S. index might be more accurate if weekly average prices were used in calculating it. The N. I. C. B. index weights were derived from the B. L. S. weighting patterns, so some of criticisms of the B. L. S. index weights also apply the N. I. C. B. index weights. The food prices for the N. I. C. B. index were also collected from chain and independent stores. In addition, the N. I. C. B. index is based on less information than the B. L. 5. index. Of the 40 cities covered in the N. I. C. B. for U. S. city average index. only four were priced monthly between 1951 and 1958. Prices for cities other than the four cities were obtained only four times a year. The three-month period index can not reflect the frequent changes in prices of foods. Comparing the A. M. S. index with the B. L. S. and N. I. C. B indexes, the living component of the A. M. S. index spreads over a much wider geographic area, with the entire state serving as a basic unit for original price estimating. Moreover, as for food coverage, only one item of frozen food, the frozen haddock, is included in the A. M. S. 1959 revised index. The use of such food is growing fast, and doubtless will be increasingly important in the rural areas. This will probably create a larger error in the food and tobacco index as farmers' use of frozen foods continues to increase. - - O'- hyt ) -‘u .....--.. ... I" .\ L ‘ ! 4. "V‘w‘ ‘V'f‘ ._ a "’ “:.1“t ant“ .J K . v- . . ~ :3 "E: f." ’he R I \ "-AI- ...“ 5.. ~ g n \— “':. ... a . ‘ — V A ‘ Knit“ He’l 'v-w O u“. . V: . , _o.._ ~ u "--..yp “F‘... a v. a. . Av- C.:l"‘u.‘\‘ ;.. (I! ‘. (2“. "o- A “ '1 I": . I t ‘ h «.e ‘i . d- ‘ . Q"\ A.‘=‘. ’ K 5 "ME L; \J I S- . - ~~‘sE A: V‘ .‘ ‘ ea“ , -g ‘. ‘ § '1 . t3 -..\ Q ‘. 4 ‘. c.- ‘\.L . x: ' ‘ 4"“. .\‘ ‘I‘U'T -. "i V « V7.9 _ ‘au; p4 ~ 1) 1 Sources and Preparation of Data Besides the M. S. U. Consumer Panel retail food price indes using panel data, two different indexes were calculated in order to compare the price movements of Lansing, Detroit and the U. 8. city average and to check the difference in the data between the panel series and the B. L. S. series. One index was calculated by using the B. L. S. value weights and price base (in the retail food price index for the U. S. city average), .and the M. S. U. Consumer Panel prices in the given period. Another was calculated by using the B. L. S. "relative importance of value weights" (in the retail food price index for Detroit), the M. S. U. Consumer Panel Prices in the given period, F” \. \ and the M. S. U. Consumer Panel 1955-1957 average price as the price ' bas ej\ The B. L. S. "relative importance of value weights" in the retail food price index for Detroit. --The "relative importance of value weights" of an index represents the value weight multiplied by relative price changes from the weight date to a later period and the result expressed as a percentage of the total for all items. 11 For a detailed description of the ”relative importance of value weights, " see Chapter II, Deriving the B. L. S. Value Weights. Because of the fact that neither the price of each item in the base period nor the index of each item in . Bureau of Labor Statistics, U.S. Department of Labor, "Relative Importance of C. P. I. Components, 1957, " Monthly Labor Review (July, 1958), pp. 767-770. ~ (.:., eLS 5' ... .... " 4 u.“‘ I. .- J ... .3 'X a .... iii-3x "I“ ’ . not . D .... ‘w-- .:;.325i '3' V-A-O». ‘ -.--A. ,“ - —t- . a .3 a... a . s. » « .CaA 10 any later date is available, the value weights for Detroit cannot be derived. Thus, the relative importance of value weights for Detroit of December, 1950 and 1952 were used as weights in the index cal- culations to correspond with the B. L. S. adjusted index and revised index. The value weights and price base of the B. L. S. retail food price index. --Through December of 1952 the B. L. S. index was calculated by using the average price of the five years, 1935-1939, as a base. The value weights were based on the 1934-1936 family expenditure pattern and adjusted in 1950 to reflect postwar changes. The B. L. S. index was then changed to the base, 1947-1949 = 100, and the value weights then represented the 1951-1952 family expenditure pattern. The B. L. S. value weights used in this study for index calculations were derived from "the relative importance of value weights" of each item in the all-items index of January, 1950 and Dec- ember, 1952, respectively. The price base was derived from the index of each food item published in the Monthly Labor Review. The M. S. U. Consumer Panel data. --The M. S. U. Consumer Panel was started in February of 1951 and was discontinued in December, 1958. From February, 1951 through December, 1958, about 250 families reported their food purchases each week. The representative sample was drawn from Lansing, Michigan, with a population of approximately 100, 000 people. This sample was selected to account for the char- acteristics of size of family, education of homemaker, age of homemaker, . . ”v.0... ~v -r~ ,,...; 1 ' .... ‘n..-‘nu V. ...e .:--ya. n:,.,., . s 6. . v. than-A0. , .. . . . .. ..- . , r. o. "c"- .... I‘M.Q\I¥‘ DU \l §~AIOAOI ...”; .,;,;.¢;‘_.;.. ~O-1h “I“ uyuxnou-n M , . fi\- - I‘ V. so. I- D - . - - a .4.»..»'.:"‘ , a... . .- ‘ . \ .tAou- ...". . ‘._. ‘ .. . _ .-~ .,, Irv ..., u.- ...C ....-_,' AD— . N —"v-‘ , I m.‘ “....“ ._r. ' . ..-- _. .7 :- ‘v-n .._‘ .- ~o ‘ ‘kL‘u N)- 9'. ...... _; - - ... ,-.~.> \. “\—"‘-a. V A - -...‘ -. - _, u... .‘ -\ ‘ 1“ \ ,.‘ -., . u a I ‘ ‘ . C y.’ ‘» ... - ". "~ u ' h “ ~ -»., , ......» - ‘Hn'\. .:. §..\_, V'- ‘7‘. a n.“ ...;p r . ‘v . :4“- ‘ U». ' ; - A v!“ s ’ a. "V , ’A‘: #. ‘n .. . ‘-. ‘F e ‘ v ‘1». ..-; ! ‘1 v... '_ ‘ “A 4‘ ”A {l' A —_‘ ‘ 11 and the amount of income received. 12 For obtaining a representative sample of families, three sample census of the Lansing population were conducted to determine its characteristics. The sample techniques and the characteristics of the sample families are discussed in greater detail by Shaffer and Moss. 13 Purchases of food reported by all families were approximately 900 items. This includes all food items that can be purchased for use at home. The information received from the panel diaries was edited, coded, and punched on I. B. M. cards. The price, quantity, expenditure, income and other data were derived by using I. B. M. equipment before this dissertation was started. The prices were the weighted averages for the week, i. e. , total expenditure divided by total quantity bought for each week. The prices used in this study were the 4-week averages and thirteen periods a year. Limitations of the panel data. Until 1953 the panel price series were only available for the items of meats, poultry and fish, dairy products, eggs, fats and oils. 12 Gerald G. Quackenbush, "Demand Analysis of the M. S. C. Consumer Panel, " Journal of Farm Economics, XXXIV, No. 3 (August, 1954), p. 417. 13James D. Shaffer, "A Plan for Sampling a Changing Population over Time, " Journal of Farm Economics, XXXIV, No. 1 (February,. 1954), pp. 153-163; and Thomas N. Moss, “Some Relation- ships of Selected Socio-Economic Factors to Food Consumption and Expenditures, Lansing, Spring, 1950" (unpublished Ph.D. dissertation, Michigan State College, 1952). ...... m... .L. ’t'n‘l .. in “a ...t‘ C ‘ . ::.e:;:; the 3):: c r a ~ A 7 355:: items were ‘ e -.... . ICI Citf‘i " .ID , “v N‘ ...:es were derive r: ' . .. :rrors Ircrn 1a!“ \..~.A\L COP 4.1,. v o ' ' ‘ “en a I 'TZ'E; 15 I a “Ed. ~ *t Pr 3*. {75 “m 12 The fruit and vegetable prices were tabulated starting in 1953. From 1955 through 1958, the panel data included a complete price series for every item in the diary. Some errors on the panel data could possibly be caused by reporting and processing. Errors in price caused by omission or by entering the purchase under the wrong heading are unlikely, since only major items were used in this study. Failure to report the actual expenditures for each item could cause the errors in weighting. Since the prices were derived from expenditure divided by quantity, any error in expenditure and quantity would affect the index. In addition to the errors from family reports, the errors caused by data processing and I. B. M. computing must also be recognized. The nonhomogeneity of products purchased by the panel families from week to week probably leads to some ”false" changes in the prices of these items due to the panel method of getting the average Price series. That is, total expenditures for a product were divided by the total quantity (including various qualities) for each week. Nevertheless, when a four-week average of the weighted average Prices is used, it provides a fairly reliable measure of the prices actually paid for a product by the panel families in that period. .2 . .. Wu. e v. v. “A .m. cc. 5‘ It I C. ..C .v s w. ... us. “a e .\ ... ... ad ‘ . . s e ‘4 . .\ .u o. o. .A“ a e .-. i.‘ ‘5‘. r n mun N‘ ken “W e . . .c d ... r V. C C ...u a a E. a r 2. 5 a... e e F. c. a .3 C C u r e 2 Mn u s .3 a. E. . A (a g r P. s s .- "B ..u C 9‘. . ‘FM nu.“ v N 3 E: «.V Hrs AM, ”a e he .. c Q e e e 1’ 1 A . 3m 5 NA an e e A s C E .. 3 .I. a 4.. . s . V. . u nhw a» . l: I: -L . Eu 8 BK Q . n...“ ‘ e o a“ -fi I‘d AV AW. . x e‘ 5 b e 11’ ‘a v“. e. ..i» .§. 0. as. .-m :5 «v ‘5 .... ‘1‘ Q» n. .... n... s. .. . «s it. a I“ .... ..J. y. n. I . w \- ... .... t I - a s . s q.» u e . \ g u-\ ..o‘ . I . e ... ..~ .... ....x » ... ... a s. .n .. I. . . u . .s \x ...oke :- eufi \\ a. s\ 5 s .‘h. i . ,u s . so. . s 13 CHAPTER II PROCEDURES AND METHODS Adjusting the Data Because of the limitations mentioned above, certain data from the panel used in this study were adjusted and corrected before they were 'used for calculating the indexes. In checking the expenditure data to be used for weighting, two comparisons were made. They were: (1) comparing the expenditures for each item between the years of 1955, 1956, and 1957, and (2) adding the expenditure for ach individual item to compare it with the expenditure for each group of those items as computed by the I. B. M. equipment. If they were not comparable, checking and correcting of mistakes in the weekly observations were made. In checking the prices of each item used in calculating the indexes, four methods were used: (1) the 4-week average prices of each item, which was selected to represent the price movements, were plotted graphically for a double check of these prices, so as to compare the regular price changes of each item from period to period within the year and to compare the prices in the same period in different years; (2) the price changes between the similar products in the same group were compared; (3) the prices of each individual item which.were observed from the panel and the B. L. S. series were compared; (4) the correlation coefficients between the prices of the same product in :;:a:.e.se:‘.es and I‘. o C i-ié"::r::‘.:'.s, 4:10. 1: 3:31.33 3 Céialmf: . . , . .. .. . ,, - ,_, urn-41. er'ci‘k.s. ...e, 4.... 9 aikyLrV“'i v 22:25 xere use: to L ape-.1345 year c: the :Iiife ‘vn-v~ .......ed 1r. only :e:::;~~- ' . , ~ . nous D\ "s:v‘~ fr 1 ‘- 5 ...e I‘:I~ _ . 'Isv.‘ . :1 ‘ i I ~ lee-‘5 Chaise: 4kgh‘ ’» . “C 1:128 ‘XICES ~,'_ i“. 2 .H\ e mmemer' 4:..2'54 "N. or HA9 tsb- aV~.3“ ._ “.c: “n.3, .‘Stezt 1P 9‘. \ ¢ ‘Q.;.' name “r A ere C 3“. . ”:eCCW~ hiDaI‘ o ‘SZI‘S l w. 1h» A‘ ‘ . ‘69 C;::'< “'13 n. RJr‘m fivSes (3‘. 14 both panel series and B. L. S. series were calculated (only cereal, bakery products, and meats were calculated). If they were not comparable, a detailed check on the original data was made for the individual products. According to the type of errors and omissions, several methods were used to correct an observation. They were: (I) To recompute if the errors were caused by miscalculation. (2) To derive an observation by using the same rate of price change in either the previous year or the year following if the prices were mistaken or were omitted in only one or two periods. (3) To derive the ob- servations by using the same rate of price change between the two previous years chained forward, or the two following years chained backward, if the prices were mistaken or omitted in more than two periods. (4) To derive observations by estimating the prices based on the price movement of similar products, if the prices were not tabulated or not available in the diary. For the products which were not consistent in qualities over time or not close to the B. L. S. speci- fications, some were combined and some were separated in order to facilitate comparisons between the panel series and the B. L. S. series. Compilation of the Indexes Three different indexes were calculated in order to serve the various purposes. Two of them were based on the B. L. S. data and the panel data. One index was based on only the panel data. The .\ V .... A; c‘uy-‘E the pl "’QF o ‘a.- _. a ‘ A. . .0 g 5‘. L’““\ r “menu-“9H. v. V. b “ ' !\ u '2"r-.. -|_\ 4....“ Dru. \ 3 ‘5 oflfi- . a» .bl.0\kec t1,“ ‘ ...E 1:.“3pge“’ ; ‘ ‘ “ ncex .‘ ‘ :‘P-P} t“\ ‘ me -_ , GHQ "a ‘L -. ‘ ~ '5 F‘" ‘ \H Da:e ‘ .. Q5 "ills. . e _ ‘ kptrlf‘h * V 15 purpose of using the panel data and B. L. S. data to compile the index was twofold: (l) to facilitate comparison of the price movement between Lansing, Detroit, and the U. S. city average; (2) to facilitate evaluation of the differences between store sample and panel sample as sources of data for constructing a food price index. The procedures and methods of constructing each of these three different indexes are described below. The first index. --This index, which was compiled by using the B. L. S. weights and base (in the retail food price index for the U. S. city average) and the M. S. U. Consumer Panel prices in the given periods, was designed to compare both levels and changes of prices between Lansing in the panel series and the U. 5. city average in the B. L. S. series. Thus, the procedures and methods used in calculating the index followed those used by the B. L. S. as closely as possible. The component indexes such as dairy products, fruits, and vegetables, ’ etc. , which measured the price changes for selected groups of food were classified in exactly the same manner as the B. L. S. index. The procedures and methods of calculating this index (using same weights and base as B. L. S. used, but using the different prices in the given periods) are summarized as follows: 1. Selecting food items. The food items used in this index calculation were based on the items priced by the B. L. S. for its retail food price index. Until the 13th period, 1952, 59 food items were selected to represent the Lansing food price movement in order to "tiff wt 'hose 5. o 1 I ‘ 0,. . .1. 1y” . .‘-C un‘ta .At . .. ‘- 1:212:31". the B. L. ‘--~ ... A. 33.”.; Ae:5\KCTt S‘f' “‘vm - - .-.t..,,-._, C... <1 i"' .P‘b>-‘.‘. V ‘ ‘ ......Cecagg‘. QC -5 ’ U .:...- , u ‘sn.‘. . ‘ h F“ .‘ \-~-\..' ~.EV'V‘\..‘S “-.‘ "QF ‘ “ , \.“ ‘A ...: ‘- -._-,u:: o A 0". ‘ u 04.. a“ q ‘ “o 1.. is. .\.v.. 33"- A' “\e V‘- -. . AC.‘ 'u. u_‘ . e ..- .- ‘- d .i: . ‘0 2‘; W h . s v AeD‘ _ i e ‘eze:-.c; _~ ‘.'_ ‘5' ~."- h§~cr ‘ ‘ 1 .Herllc _ ". ...»- u...- x». .\ .. ... T1.“ “‘5 r»- I ‘I‘l ~_ F. : *xfe ‘ 16 compare with those B. L. S. adjusted index. From the first period, 1953 through the 13th period, 1958, 81 food items were priced to compare with the B. L. S. revised index. However, the complete index for all food groups was not started until 1955 since data on some food groups were not available in the panel. In order to keep the panel food items as close to the B. L. S. specifications as possible, some necessary adjustments were made in the panel basic data. For example, the panel data for broilers and fryers combined were sub— stituted for the B. L. S. data for fryers only. For a comparison of different tiems‘ between the B. L. S. and the panel series, see Chapter V, Table 9. 2. Deriving the B. L. 5. price base. The price base was derived from the B. L. S. index. The index of each food item is simply a price relative which represents the relative price change from the base period to a later date. Symbolically, Pta Ia = ——- 100 p0 a ,, . . . . . Where "Ia represents index of item "a, " "pta" is the price of item "a” in the given period, and “poa” refers to price of item "a'I in the base period. Thus , O _ a.100 ptal and 1a were published in the B. L. 5. Monthly Labor Review. 3. Deriving the B. L. S. Value Weights. The value weights .3 . . .- _ Z; .L ...... . r“ ..u ..l\ x s n . o . e ... t A tr“ 6 8 m1. 3 .t X . n m: V . ac . A ..A r 0 FA ab . r“ t. 8 .. .. . 6 D. H ‘4 .PAM s .9» v. .t 0. Cu :15 e 9. . “P” ”in “ “a. m “.4 r v O «.4 ,. . ..k. x.“ p .1 I]. and!“ .BW Va . V. rJ. e 9». o . ,v. ..flh.” 0&4 M“ 'A ‘5 “J‘s ' A r. u 0 Q ‘ .‘ a .\ ....u . a HP; G. on 9 . VP» t o a . s A n . o urn . H kw FL ..n . us... . 3 ...n n u . . e. u u: at. . - ~ ”.n .1 .- n ...a .... .. .. a... a . ..... a... .. .. ‘U. I. K M .II Q V~k~ O s: D. s Q »L e \A e ..u o . .‘v LW 17 were derived from the B. L. S. "relative importance of value weights” in the food price index. According to the definition, the B. L. S. "relative importance of value weights” can be written Where CPI represents the consumer price index for all items, "RIa" means the relative importance of value weight of item "a" in the CPI, the subscript "wa” refers to the value weight of item "a" in the CPI, p03. and pta, as before, stand for the price of item "a" in the base period and given period. Transposing the above equation, we obtain w . — . 100 = RI . CPI p a Thus, a a pt . 100 The value weight of each item was derived by using this equation, i. e. , its relative importance times the ”all item index" and times the reciprocal of its price relative divided by 100. 4. The index formula. The basis of the index was Laspeyres' formula, v m.. , war: . .‘\‘CD~ o l . . :I.-‘..":V>" ‘ a... .v-~,.” “‘ tn fi I . ...-a :1.qu\ . ”A II "U\- Y n ‘ 'k \-.:.. v- s. _ . . V“. “t: {- ‘Kr- \ ‘ . fi . . ;‘. ' — u ‘ v ". -‘X.‘_ ~_,{-v~_‘. ‘ ”a. s ‘\ t .1 ‘ t F U‘ v ‘Q h- “‘1‘ ‘s_ r. .. l8 zqopt It = -——— 0r, qupe pt qupo 3- It = o {(10% Where It means price index for a given period, ”p" refers to the price of an individual product, the subscript ”0" refers to the base period from which price changes are measured, and the subscript "t" refers to given period which, in this study, refers to M. S. U. consumer panel price (4-week average) which is being compared with the B. L. S. base price. In actual practice, however, a variation of this formula is used, namely, 1:1 EfiL— 01‘ t t-lzqopt-l :ex 1: I=I -—-—— t t-l {ext-1 The last formula is the one used, where ”ex" represents the individual item expenditures with base period quantities, "t" for the given period, and t - 1 for the last previous period. The summations are performed over all the items in the group under consideration. The first step of the computing procedure is to get the current item price relatives "Rt". The current item expenditures are ‘ r ‘ ,. ‘U' ' .. .. V'HH .' _.;-. :u malrevA““L ‘ ’.....o mitt 'zlx' 1‘70: 5“ '14 p~¢ ‘ A. v a .mnunsbfl“\| .:. -... I §. m -7 -~-.u:C:. ate-Ac exvt O ... I - , IF ‘ ".:.: Le‘u ?'..; 3"" . ‘ by ‘¢‘ “dob. TV A r. :‘9 Any.» «.32. “h". “v- ‘ . _ ‘- fl 0 . ..C a. " AM Lon‘ '.. _.~.~._ . a... . “" . -.... dcr‘r‘m . Ash) I a 7-- q .4. 1*e ‘V‘ as ‘ ’- .. ‘0“... ..‘ K :c ;O-- ‘ - 5 fl "‘ . V v . ogg‘i‘c ‘v- —.. - .....,.. .m" ... '1‘.‘ e. . . .. 9 k. ‘rlfie - Xe: ~- ".‘ ..;.- e 5 ~A w it“ $1, \ w . o‘.e :'F~ a‘_.‘u'\ v bv’ ‘ ... C ., I. ‘ «t "A yak \‘T‘ :1 ‘ 6 Q ‘~ ‘ a I ~\. . .-\ ;- ‘ ~-:“ n. - . " We'm ‘ ems - t; -_t .A‘ . “Jr; 5‘. {RIC a e ). ‘ . E . . -~', “1,. ‘\..‘ ‘ , a " ‘ .""t "‘ we ‘ .1: . 19 obtained by multiplying previous item expenditures by price relative, then divided by 100, since Rt 2 t . 100 pt-l Symbolically, ex . R t-l t ex :- t 100 The current item expenditures are totaled at the categories, subgroups, groups and total, and indexes are computed at each level by using the index formula. That is, multiply the expenditure ratio between the current period and the last previous period by the index of the last previous period. 5. The linking procedure. The linking procedure was based on the standard linking procedure, i. e., price changes from the 13th period, 1952 forward were weighted with revised weights and chained to the index of the 13th period, 1952 computed with old weights to obtain the indexes for subsequent periods. The first step of chaining was to get the link period index (in this case, the 13th period, 1952) with the old value weights; the second step was to set up a second set of revised weights for the link period; and the third step was to compute the period price relative from the link period based on revised weights and to multiply the previous period's index (the "lead on'I period, in this case, is the 13th period, 1952) to obtain the subsequent index numbers. For example, let It 1 = 105, where It 1is the same index .....- v ‘ ..5 I. \- .. ~p§n . _..,..‘.. ~"v—‘4' "kn-I\ A. V‘ ‘ n h ..c “ 20 of the 13th period, 1952, calculated by using old weights. Let ql, q2, q3, . . . q , be a set of new quantity weights n for items 1, 2, 3, . . . n; pl, p2, p3, . . . p , be the prices for item n l, 2, 3, . . . n in the 13th period, 1952. The price times the quantity, qlp1 + quz + q3p3 + . . . + qnpn 2 ext_ 1' which is the sum of the expenditures for all items in the 13th period of 1952. The sum of the expenditures for all items in the lst period, 1953, is qlp'1 + qu'z + q3p°3 + . . . + q' p‘ 2 ext, where p'l, p'z, p'3 are the prices for items 1, 2, 3, . . . nin the lst period, 1953. Suppose ext 1 = $200,000 and ext = $220,000. Then, the weighted price relative from the link period to the lst period of 1953 is The index (It) of the lst period, 1953 is 105 x 1.1 = 115. 5. If the sum of expenditures of the 2nd period of 1953 (ext+1) is 224, 400, then, the _ 224,400 _ t+1 ‘ 220,000 ‘ 2nd period, 1953(It+1) is 115.5 x 1.02 = 117.81. The subsequent weightedprice relatives of R 1. 02. The index of the indexes I . , 9 I . . . I are obtained by the same procedure. t+ d. t+ 3 t+n Only two group indexes--"dairy products, " "meat, poultry and fish"-- were linked. The other groups, which were started either in 1953 or 1955, did not require any linking. ' 6. Seasonal food. Items of seasonal food were priced in the same manner as the M. S. U. Consumer Panel retail food price index, 2" v- o ~b~m..v‘.’ 21 which will be discussed in the latter part of this chapter. The second index. --This index, which was compiled by using the B. L. S. "relative importance of value weights" (in the retail food price index for Detroit) and the M. S. U. Consumer Panel average 1955-57 price as base and the M. S. U. Consumer Panel prices in the given periods, was designed to compare the price changes between Lansing in the panel series and Detroit in the B. L. S. series. In this index calculation, the food items, index formula, and the linking procedures were the same as those used in the index based on the U. S. city average which were described above. The methods of ”linking, ” "conversion of base, " and “twelve months compared with 13 periods" used in this index are summarized as follows: 1. The linking procedure. From the fourth period of 1951 to the 13th period, 1952, the indexes were weighted according to the B. L. S. ”relative importance of value weights" of December, 1950. From the first period of 1953 through the 13th period, 1958, the index weights were based on the "relative importance of value weights" of December, 1952. As in the case of the U. S. city average, the 13th period of 1952 was used (as the link between the indexes which were calculated in different weights. Since the B. L. S. base price for Detroit was not available, the M. S. U. Consumer Panel average 1955- 57 price was used in order to compare it with the M. S. U. Consumer Panel retail food price index. 2. Conversion of base. In order to facilitate comparison . .‘. .9 ..II . ‘ .:ae .f..5 Air“ ass I. ”tn-R :zriexesr “51*" Y -.S. a'~€-’ ’1 ll’ 0' £6 CZ 11:..atei $5155 13‘ 5, Twelve m; :5sz were calculate :ite Lndexes 35mg “2:: : ‘Ncfl ..e.,13_:eri: K . yeleCS ‘I.‘.‘ 61:“ ,. ' . 36....28 C1352. A :;4 §_ ' l " P F4 “m .g';re 3 \E '- e 6“ ' ‘ ‘. \vlaec‘ec Ch :~-. \:::;_':E .7. \ V "‘e ‘dacexe 2:?-:y ‘ l-r“ . nae m‘jh§: ‘ ‘n 1 I.“ . M. N“. .L 22 between the two price series, the B. L. S. index (1947-49 = 100), except in Chapter IiI, was converted to the same base as the M. S. U. Consumer Panel food price index (1955-57 = 100). The procedure of conversion is to multiply each B. L. S. index by a conversion factor. In this case, this factor is the ratio of the average of the three years 1955—57 indexes, which are calculated by using 1955-57 = 100, to the B. L. S. average of the three years 1955-57 indexes, which are calculated by using 1947-49 = 100. 3. Twelve months, compared with 13 periods. The B. L. S. indexes were calculated on a monthly basis, i. e. , 12 months a year, and the indexes using panel prices were calculated on a 4-week period basis, i. e. , 13 periods a year. Either adjusting the B. L. S. 12 months index to the 13 periods or adjusting the 13 periods index to the 12 months would change the values of the index. Since the comparisons were made mainly to measure the movement of two price series over time, the 12 periods of the B. L. S. monthly indexes were plotted against the 13 periods of panel 4-week indexes in each year as in Figures 1 to 14, and "Figure 31. Because the prices used in the B. L. S. index are only collected once around the middle of each month, one way of comparing the indexes between each 4-week period and each month is to compare the month with that period which includes the 15th of that month. In this case, the first five periods can be compared with the first five months and the last five periods can be compared with the last five months, in each year of the eight years, since the 15th of q I “ .- 11217.3. 161; 11‘. est. "‘ " R. . .. n " :::.e .1326 5.5. 2126 average 0: se azure: are. \l. u,- u- ‘0“. , ’ F".-§ . . . .-..» 50:. D. ‘Chr‘s ..r L‘H'f- , . ""ny fiv- I .. v m .. uH‘Er 23 each month fell in each of these periods. And June is compared with the average of the sixth and seventh periods and July is compared with the average of seventh and eighth periods in each year of the eight years. The third index. --The retail food price index of the M. S. U. Consumer Panel, was designed to measure the effect of price changes on the cost of foods in the "market basket" in Lansing. The index is based on prices actually paid by the panel families. The average size in 1955—57 on the families. included in the index was estimated to be about 3. 2 persons, and their average family income was estimated at about $5, 452 after taxes. The procedures and methods of calculating the M. S. U. Consumer Panel retail food price index are summarized as follows: 1. Weighting and selecting food items. The content of the "market basket"--that is, the quantities. and qualities of food that represent what families bought in 1955-57--is assumed to remain the same, so that the change in cost from period to period is the result of changes in prices. Sampling assumes that a relatively small number of properly selected items will be representative of an entire group. For this purpose, the data were first arranged according to the major groups of family spending on food. The five major groups were further divided into 20 subgroups (see Table 9, Chapter V). Thus, a definite stratification was established in which each of the basic expenditure items was classified by type and assigned to a proper (subgroup. Wt) was 1;: ' . a ......US - A -~.~ ":e fizzle cue -; -. .326 "‘A‘n ' sn“-.. \. -.. ,; -g. ..h‘vnsnEM ei\‘. 4-\‘t' ‘=‘*: :. . . ..-. .....u.» 9X: 3'- ‘7‘ . ... .,. .'_ r» ‘1 ‘_ (I) ‘U ‘ Q i a I-.'=~‘ A. ... v e ~— Oc. O 5.4": . ‘~ - . n .. ~ g. - A . F . ‘ ...S.CE‘ ; ‘ in Z" "n -: ‘ ‘ s.e“: e‘A‘u 1 1‘ EN; t..‘ 24 Each subgroup was further divided into small subgroups and categories. It would be time consuming and costly to analyze the 4-week price data for all nine hundred basic expenditure items. The data, therefore, were further condensed within the structural framework and a smaller list of items was selected to represent basic expenditures. The main criteria used in selecting the items, which were to be priced each 4-week period, were: (1) Importance of the item in the total family expenditure. The effect of the price change of any item on the total price change varies with the relative importance of the item. (2) Representativeness of the item in a group of items. Since the price change of the selected item will represent the price change in the unpriced items in the index, two factors were considered in determining expected similarities in price fluctuations. One is the pattern of the past price fluctuations, and the other is the usage of the products. The basic expenditure items were grouped according to the above considerations, and one or more products were chosen to represent each smallest group. The expenditure for the selected items was then used to represent the combined expenditure of the group it represented. In order to clarify this procedure, let us take the fats and oils group, for an example, and follow, step by step, the method developed for this particular case. The first step in the development of the market basket for fats and oils products is the grouping of the primary data. Assume, ’1 o .... ~|\ 'vu Q 1‘ w‘. I ('9 ..- ‘116 i ‘ ,. ‘u ‘ . .. . . 1.: 516 iSSamEC . _- ».. ~. ~ ’- V‘ i“ v<“,. . "=¥" :hK-‘Ageuc - " I! ‘ ' A"“ "I- .‘V § ' ':K7-=3-€ -..J., . 1 ' \t-..o‘.'~ . _* “‘IA‘0¢K \— . tz‘fl \ -’ A. "4» .;.s (. 3. ., ._ ' ‘. P~ ca. "‘Vl, t3.- '7“ I Aue a -"\s '. 3,. w. ‘o‘ . Ngu¥m as .r‘“ ‘V. U l. 0. A\ §eyn L». ‘ .2 .‘s‘ . ‘7 e o. .:‘e 4 . \;~.'_. “ n.- '~ 3P- ‘\¥::r:‘y- ‘~‘ 1:. ‘ “I u e u -. ‘- .:S 25 for purposes of this example, that the total amount spend of the fats and oils products group represents 100 percent of all expenditure in the food group, and this breaks down into the 11 items listed below, l which are assumed to be purchased in the proportion given. Item Percent Item Percent l. Oleomargarine 37. 51 6. Lard 2. 57 2. Vegetable shortening 16. 81 7. Other fats . 40 3. Swiftning 2. 32 8. Peanut butter 17. 18 4. Salad dressing (10. 64), 9. Mayonnaise 4. 56 french dressing (1. 90), roquefort dressing (. 40), 10. Sandwich spreads . 48 salad dressing mix (.05), diet dressing (.13), ll. Whips . 32 total 13. 12 5. Cooking oils (3. 68), salad oils (. 59), other oils (. 46), total 4. 73 The next step is to select from this list the representative items which would priced for an index and would constitute the "market basket. " According to the criterion mentioned earlier, we consider each item as follows: (1) Oleomargarine. This item occupies by far the most impor— tant place in the fats and oils products group and should, therefore, be included according to the first criterion given above. Using the percentage Butter was classified in the ”dairy products" group instead of the fats and oils group. . - o O .. .n‘. t ...... .... em 3116; I u - - ~o,‘..‘ . ,. ... .. G :Ctdbl‘ CA\ u. 'r ..ElZiIiiIlEC 1'15 8 5 .J ....... o n --..—..n v-' -‘ 1:115 pf.L€ 1 ,~.: -v \HVAH 4.531611 :mT-J‘: I: w- ...x...:):100,1 I t . C3mCaIL fh'L... ~ a... ...,Vt d price Cr n'f: h .' . ‘ u g. “:6 :ten" '7- an. . ‘r-r: ‘ ' vb¢5y\‘ .:Eit Iii, 5‘ v‘ ' -. .._ : . 'u' n sgvc EX“ '- c. J lClI‘Te‘ :91 ‘ n. J. C :36'p'” . A ‘ \'erlt C): a: e~ .g. ‘ 5€tc3.e {a (j S .4 v ." *kitr‘z‘g ‘ 0 he . N_;E- 5‘ iv "'a%‘ A '5 h ‘ auht l::::; I‘ HEMLHS ‘0 ~ Coy-‘5... , - U '~. "4 5. lie I. 558‘ I , ‘ l 0‘ a W“: 1’. 'iIA 26 assigned to Oleomargarine in the total expenditure weight (37. 51), it can now be easily illustrated how a change in the price of a relatively important item affects the index as compared with a change in the price of a relatively unimportant item. Assume that the price of Oleomargarine rises by 5 percent with all other prices remaining constant. This price change is reflected by the fats and oils products index as 100 + (37. 53 x . 05) = 101.19, or an effective increase of 1. 19 percent. On the other hand, a similar price rise in Swiftning, an item of small importance (2. 38%), will be reflected as 100 + (2. 38 x . 05) = 100.1, or an effective increase of 0. 1 percent. A comparison of these two index figures demonstrates clearly how a price change is reflected according to the relative impor- tance of the items. In order to see their true representatives it should be noticed that fats and oils products form only a small part of the total food expenditure, of which the entire group represents only about 3. 6 percent of all food expenditures (see Table 9, Chapter V). (2) Vegetable shortening. (3) Swiftning. The relative importance of Swiftning is not enough to warrant individual pricing. It can be combined with vegetable shortening to constitute one subgroup since their usage is a similar one. The item with the greatest relative importance is vegetable shortening in this case, and it is selected to represent this entire subgroup. (4) Salad dressing, french dressing, roquefort dressing, . V ' '_ ., 'I'E'SS‘ong 1.5.x: , . ._‘_ .A . 4‘. ‘ I. K r in: u... as -‘ . pry-rue ... U ‘ ,...-.:;:e 0: ::e s; -5:':~... S‘VHP. 5; p emit b‘ . H '21:. ‘9: Lll't. VI'Ler s‘. COWS ‘ t.» ‘1 27 salad dressing mix, and diet dressing. They have a weight of 13. 23 percent and will be priced as a group for the same reasons that oleo- margarine was priced individually. The item to be priced should be one which has an established market, and the price of which can be expected to fluctuate in the same manner as the price of other dressing. For these reasons, salad dressing should be a good choice. Further- more, salad dressing has the greatest relative importance in the total expenditure of the subgroup. (5) Cooking oils, salad oils, and other oils. The expenditure weights of these four items again are of a magnitude which does not warrant their being priced individually. They are all items of a similar usage. When they are combined they form a sufficiently important group for pricing. The item with the greatest relative importance is cooking oils in this case, and it is selected to represent the entire subgroup. (6) Lard. (7) Other fats. Lard and other fats are the items of a similar usage. They are combined to form a group for pricing. The item with greatest relative importance is lard. Therefore, lard is selected to represent the subgroup. (8) Peanut butter. The proportion of expenditures made for peanut butter shows that it is a fairly important item. Furthermore, no other item, according to the criterion of suitability to represent a 3;; 5:. be combzm Lie agrees to Star: 11 0 ”'. v 6 U. (h n U :1 (I) .. n. ’1- 91 F . . EinCWLC These 1: L‘Izltiroug} 0 b” ... ‘ ’9 M - .:ex expenditur e \k e 28 group, can be combined with peanut butter. Therefore, peanut butter, will be allowed to stand by itself. (9) Mayonnaise. Mayonnaise with its weight of 4. 56 is important enough to be considered alone. (10) Sandwich spreads. (ll) Whips. These items do not constitute one group as do the items from 1 through 9, but are a collection of miscellaneous fats and oils. Their expenditure weights do not warrant enough to form a group of their own. It is logical to assume that these items will fluctuate in price as all other fats and oils combined. Therefore, their expenditure weights will be ascribed proportionately to each of the above subgroups. After combining the expenditure weights according to the above considerations, we finally arrive at the following market basket for the fats and oils group: Item Percent Oleomargarine 37. 81 Vegetable shortening 19. 28 Salad dressing 13. 23 Cooking oils 4. 77 Lard 3. 00 Peanut butter 17. 32 Mayonnaise 4. 60 Total 100. 00 As was already pointed out, this example is a small and simplified model illustrating the derivation of the priced ”market basket" from the panel. Hw.‘ .‘- nu. 29 It is clear that the distributions of expenditures among individual items (the price change of each item represents the price change of a small category of food items) included in the smallest subgroups of foods were based on the average ratios of item to the subgroup expenditures. Thus, the distributions of expenditures to each of the subgroups included in the major groups were based on the average ratios of subgroup expenditures to major group expenditures. Finally, the distributions of expenditures to each of the major groups included in the total food expenditures were based on the average ratios of major groups expenditures to the total food expenditures. A concrete example may help to explain this procedure. According to the panel, the family expenditure pattern, during the period from 1955 to 1957, was determined by a detailed study. The distribution of the total food expenditures to each of the five major groups was described in Table 9, Chapter V. The expenditure for the major group of "meat, poultry and fish" was 30. 7 percent of the total food expenditures. The expenditure for the subgroup of meat was 71. 9 percent of the expenditures of the major group of "meat, poultry and fish. " The remaining 28. 1 percent was distributed to the poultry and fish subgroups. The expenditure for the small sub- group of pork was 32. 3 percent of the expenditures for the meat subgroup. The remaining 67. 7 percent was distributed to the small beef, other meats, veal and lamb subgroups. The expenditure for pork was 32. 3 percent of the expenditures for the meat subgroup. -~ ...4 w-‘. . a _‘ n. .. '. u ¥‘ ~. '.\ . l -" ~ '-.~. ‘I.. ‘ ‘. "'7 g .‘ ~-._‘ ~ :- I ‘3._ » 30 The remaining 67. ‘7 percent was distributed to the small beef, other meats, veal and lamb subgroups. The expenditure for pork was dis- tributed to each of the smallest categories which are represented by the individual items. For example, fresh pork roast is selected to represent the category of all pork roasts. The model illustrated above is to explain, in a general way, the procedure used in developing an index ”market basket" from the panel expenditure data, which represented an actual record of the expenditures for foods purchased by all Lansing urban families at a definite time. It should be clear that although every item is not individually priced, it is represented in the index. 2. The formula used in calculating the M. S. U. Consumer Panel retail food price index. The method of combining prices into a price index is basically the technique defined by the Laspeyres‘ formula. The simple mathematical notation for the formula was described earlier in this chapter. Since the panel index is based on the panel family expenditures of the 1955-1957 average, they represent recent spending patterns of Lansing urban families. No consideration needs necessarily to be taken of basic changes in items and weights. Therefore, the index did not require any linking. Furthermore, the price base period and the weight base period (1955-57) of the panel index are the same. Thus, the panel index adopted an aggregative index in measuring the changing value of a fixed aggregate of food items from weight date to any other period. These indexes were calculated by the Laspeyres' formula to ., - . - \ ur- .~. .v“ u . -. .. ‘ :- ~\~.‘..-. ‘v‘ ... ‘ " \r- “ A. ..~_ . u-._ ’ . - l, H:- u._ ‘." ‘ n u ‘M u ‘ .. I ‘ll= ' I \o H. O Q. in. "u fi 7. 31 compare the current period with the base period rather than the B. L. S. chain index for month-to-month comparisons. In actual practice, the panel index used a different formula from the B. L. S. index. The B. L. S. indexes are the average relative chain indexes. Methods, of calculating the chain indexes used by B. L. S. , are the same as they are used in calculating the other two indexes using both the B. L. S. data and the panel data, which were described above. Now a more precise statement of how the formula can be applied to the panel index shall be described. The panel food price index used a ”market basket" of average expenditures from 1955 to 1957. The fixed quantities, or qo's of the formula, were derived from a consumer expenditure study based on the panel data (see Chapter V). The base period or time "0" was 1955 to 1957. Therefore, the index for any given period, say period ”t" would be written: qu955-57pt . 100 I : zq1955-57pl9ss-s7 t In words, this says: The index for period "t" equals the sum (2) of the products of the average annual quantities (q1955 57) of the items (fixed according to the 1955-57 purchases of the panel families) and the average prices (pt) of the items for time "t" divided by the sum (2) of the products of the same average annual quantities (q1955 57) and the average annual prices (p ) of the same items for the base 1955-57 period 1955-57. Since the index is a percentage measure, the quotient of the two sums is multiplied by 100 to give an index number. This . 5 . ... a x I" t“ N I h . . I . ... . . ... p. .- n a l . n . .ah at t . a .. . .. u v . J l .- 5. . . a . . ~. ....» .‘u u L» 1.. a e a u .:M u . ‘u ‘N vi ~rhv u . .an 3P. .... . ... u .. . u ..l- s s n!- \.|.. v . u . . .9. . t ..u \u it ... . . v- . 1th.. . DO. - \ A . A , av 32 formula is also used for the major group, the subgroup, or the small subgroup, index. The panel index was calculated using a variation of the formula as a weighted average of price relatives for each item: p Z ——-t——— ' p1955-57q1955-57 p1955-57 :Pl9ss—57q1955-57 Thus, the weights used in the calculation of the index are "value weights" representing the cost of 1955-57 quantities at the cur- rent prices. The summation of value weights for groups of items in this manner makes it impossible to identify the quantity factors attached to each index item despite the price base and weight base in the same years. They are of differing kinds and quantities and cannot be described in the same quantity units. Quantity weights, therefore, are only implicit in the index structure. 3. Prices of the items used in the panel index calculation. Using the information of the expenditure data, 131 food items were selected and priced in each period. These items can be used together to estimate the average change in prices of all items of food. These items are of outstanding importance in family purchases, so that the items in themselves represent the greater part of family spending. (See Table 9, Chapter V for the number of items priced for the index. ) Prices of these items were averaged weekly. And, then, these weekly averages were further averaged for every four-week period. 33 4. Calculation of the index. The average prices of each four- week period were compared with the prices obtained during the base period--the 1955-57 average. Thus, the percentage of the price of each item was calculated. These price relatives are expressed, not in dollars and cents, but as a percentage relative to the price in the base period. Then, these price relatives are put together to determine what happened to the prices on the average. To do this, each price relative is applied to its index weight to determine how much the price changes affect the total food expenditures. The following example illustrates how this is done for items of fats and oils. 1955-57 Price The first period, 1958 Item expenditure relatives expenditure weight weight after price change Oleomargarine 37.81 101.2 38.26 (37.81 x 1.012) Vegetable shortening 19. 28 101. 0 19. 47 (19. 28 x 1. 010) Salad dressing 13.23 105.1 13.91 (13.23x 1.051) Cooking oils 4. 77 101. 6 4. 85 (4. 77 x 1. 016) Lard 3.00 115.6 3.48 (3.00x 1.156) Peanut butter 17. 32 96.1 16. 65 (17. 32 x . 961) Mayonnaise 4. 60 101. 4 4. 67 (4. 60 x 1. 014) 100. 00 101. 29 We say, then, that fats and oils prices increased 1. 29 percent, on the average, from the 1955-57 average to the first period of 19 58. After similar calculations are made for all food items in the A..- a .. -. 2. .-- . '- ... “.’"‘. \. -.. n "e e. . IA. 34 index, the expenditure weights are added and compared, the results measure the average price change on all foods, and an index number for all foods is calculated. 5. Selection of base. The period 1955 through 1957 was used as a price base, while the 1955-57 average expenditures were used for weights in the index. According to the expenditure study based on the panel data, the family income, the pattern of consumption, and the prices of various products were all normal during this period. For the average of the three years 1955-57 price of each item, see Appendix 2. 6. Seasonal food. The weights used to measure average price change for certain items such as fruits and vegetables were the constant annual weights. Lack of prices for given items in given seasons was handled by estimation--either by holding constant until prices again became available, or by assuming the same price change for out-of season items as for year-round items. For example, fresh peaches are a seasonal item and prices are not available at the periods of out- of-season. In order to supply a peach price for such periods, the prices of peaches were assumed to vary in the same manner as apples, which are available at all times of the year. For a detailed description of handling seasonal foods, see Appendix 3. 7. Relative importance of items. "Relative importance" of the panel index refers to the percentage distrubution of the ”value weights” which enter into the index calculation. The relative importance 35 figures for the base period of the index represent the distribution of family expenditures for the period of 1955 to 1957. To exemplify, if 20 percent of consumers' expenditures were allocated to dairy products and 10 percent to bakery products, 20 percent and 10 percent would represent relative importances of these groups. In subsequent periods, the relative importances do not reflect the distribution of actual expen- ditures. Instead, they are the percentage distribution of the costs necessary to purchase in the current period, the same quantity and quality of foods purchased in 19 55 to 1957. They are, therefore, affected by the size of the base period expenditure, as well as by the differences in the rates of changes for prices for different items: e. g. , relative importance will increase for those items which rise in price faster than average and decline for those items which increase less than average. For example, assume an index made up of only two items, for which the expenditures initially were $60 and $40, and which consequently had a relative importance of 60 and 40 percent, respectively. If the price of the first item doubled and the second advanced 50 percent by some later date, the index value weights would become $120 and $60, and their relative importance 67 (120/ 180) and 33(60/ 180) percent. The relative importance of each of the items was calculated for the 13th period of each year from 1955 to 1958. The formula used in calculating the relative importances is the same as the B. L. S. used. That is, 36 pt pt pt w . a W b w n a — b -— n — p0 pOb pOn ma: a . loo, RIb= 100, , run = 100 R Rt Rt t C‘ Where wa, w wn are the value weights for each item in the index, b, s o 0 Pt . pt . . . . pt are the average price for these items in the 13th a b n period of each year; po , po , . . . po are the average prices for a b n these items in the base period (1955-57); Rt is the index of all items of foods in the current period, the 13th period of each year; and Rla, Rlb, . . . R111 are the relative importance for each individual item in the index of all items of foods. (See Table 9, Chapter V for the relative importance for each item. ) The method for calculating the relative importance of each major group and subgroup of food is to multiply the base period (1955-57) group relative importance by the corresponding index of the group for the period desired, in this case the 13th period of each year from 1955 to 1958. i"~ N...‘ b u 0“. _ ..‘...= cl! (0 -. . ' .:E: . .e I: . ‘5- n,‘ u u div 0 .‘ e s. .‘c. C u L ~.e 1 4 37 CHAPTER III COMPARISONS BETWEEN THE M. S. U. CONSUMER PANEL SERIES AND THE B. L. 5. SERIES FOR U. S. CITY AVERAGE The difference in the food price level between Lansing and the U. S. city average was measured by the index calculated by using the same weights and price base as the B. L. S. used, but using the panel prices in the given periods. If each item included in the index is specified as the same quality and quantity, i. e. , the variations of the indexes show only price differences and do not vary due to other factors, the two indexes would provide a direct comparison of the price levels between Lansing and the United States. However, the prices used for the indexes prepared by the B. L. S. are collected by direct pricing in the various food outlets included in its sample and are for particular specifications of each food item. They are not directly comparable with the M. S. U. Consumer Panel prices which are paid by the panel families for items of varying sizes and qualities. For example, the price of pork chops can vary considerably, depending upon how near the center of the loin they are cut. The B. L. S. speci- fication for pork chops priced: Description: Pork chops, cut from center of loins Grade: No. 1 grade Unit: One pound Exclude: Rib end or shoulder end cut chops, soft or oily pork. 38 The prices of pork chops of the M. S. U. Consumer Panel are the weighted average prices paid by the panel families during the week for all pork chops. A comparison of the B. L. 5. food price indexes for the U. S. city average with the indexes calculated by using M. S. U. Consumer Panel prices in the given periods (the same value weights and price base as B. L. S. being used) does not show exactly whether prices are higher or lower in Lansing than the U. S. city average since, in fact, the variations between these two indexes are not only caused by the differences in prices but also reflect the differences in quality. Thus, the following comparison of the two price levels was made in order to compare the differences between the B. L. S. store sample and the panel sample. However, the indexes provide a direct comparison of the average price changes between Lansing and the U. S. city average from one period to another. The Levels and Changes of the Two Price Series The panel price index in the "meats, poultry and fish group" was lower than the B. L. S. index in every 4-week period over eight years. The B. L. S. index among the time periods varied from 11. 3 to 19. 3 points higher than the panel index (Figure 1). The differences between the two indexes indicate that the B. L. S. average price ranged from 11. 4 to 23. 3 percent higher than the panel average 1Bureau of Labor Statistics, U. S. Department of Labor, Food Manual (Revised February 2,1953), p.11. p.-.' n».- v‘. rh-‘vs u I\. w "e '- ... 'N N 5" 39 price. Higher prices of steak, roast, and bacon in the B. L. S. series accounted for most of the differences. Comparing the annual average indexes, the B. L. 5. series varied from 13. 0 to 18. 4 points above the panel series. This indicates that the B. L. S. annual average price was from 12. 5 to 19. 0 percent higher than the panel series. Over the eight-year period, the mean of panel average prices was 16. 3 percent below the B. L. S. mean. The two price series had similar patterns of seasonal variations and trend over the eight-year period. Variations followed a rather regular seasonal pattern. Prices of both series were highest during July, August, and September, and lowest during the winter. This, seasonal movement resulted chiefly from meat price variations. Both series followed a downward trend from 1951 to 1954, then leveled off during the first nine months of 1955, and went upward from February, 1956 through the end of 1958, after having dropped to a seasonal low in the winter of 1955 (Figure 1). The panel price index in the dairy products group was lower than the B. L. 5. index in every 4-week period over eight years. The B. L. S. index varied from 1. 5 to 15. 5 points above the panel index on a monthly basis (Figure 2), and 2. 7 to 13. 1 points according to their annual average basis (Figure ‘7). The differences between the two indexes indicate that the B. L. 8. average price was from 1. 4 to 15. 6 percent higher than the panel average price on a monthly basis, .hbsuqanvn- .IsICx in. Iibuhh dulsCl have II-ensu tflttu-l \UWu l'hiIIN 40 onaa snag onoa nnma — -_ _ _ q. _ _ _ __ _ q _ q q. . q __ _ — a —4,4 1 .q . _ __ a a ._ . . . __ . _ a. q oh I O I u. s O 1 cm .. .. .. 18 s s s .s I s s ‘.\.,-‘.'-'.’ s s s s o ‘u ’0’ o‘- ul- d s4 \ ‘1‘ u‘n‘ ~.‘ Ila: Sfi ’ ‘ ( ’u'.‘o'a-u’-nl \. ..l \ .\. .r.t \ 1. odd o:l..'.’ \.\ \. I. .\ t l. owg t1 and «non nnou «mom anou on _ _ _ _ _ _ _ _ _ _ _ _ _ q _ . _ a _ _ _ _ _ ~,— _ _ _ _ a _ a _ _ a . .mo. _ . . q . . . _ . ..4 vauusu so>ao snu ea e «um assay uses-sou .xczn. .:. m. a van .s-sn souum was sunnus: .m. 4.. uses: ecu-asouso .n-«n use .huuasom .1 cm ::.3 ....: 2.. .z. .nusox we ssouum gunned new woven s .I I I o s. «s 1 0 4.0.10 \ OOH ...! u s s u s s o s o s ... s.‘s\ .57.}. \.J .,.l.l o 0 find... \.'.’.\- ..l .\.\ .f.. I Odd .r‘.‘ Isl ‘0'. ‘ ole c‘n‘ I.‘.-ou'a'o o.‘a-s-s‘a\u - l.\ sv.l.\ IL ONA ooauaeunead swsuo>< muao .m.= .asun vs. \\\n .auuasom .sussl no nsoqum gassed new Mecca .».A.n and uses" omens>s huuu .m.= new .b.a.n can snagged: .and-=1; new ass-m sols-soc .:.».x essays; scum use .huuusoa ..ussl mo ssouua unseen sq sons-Jo sea as ss-«usaloo .u .uah 41 cued snag ona~ anon _ .ln _ _ _ L _ _ _ _ _ _ _ _ _ _ _ a _ _ _ a _ _ _ _ a _ _ _ a a _ _ _ _ _ _ z _ n _ _ _ _ a a as {.00 ll oa Ilse-becso.u.s es...- sss .8‘ \.D..f. ‘3’ -..! -...-. I ‘ \ -.\.§. 1 Odd 0.-.-n‘.\. " I, l. 9“ :1 can .1 and «non nnaa «nod anon on (a a. _ L ._ L __ _ _ _ _ _ _. _ a __ _ a _ _ __ a _ __ a _ __ _ _ __ _ __ _ __ _ a __ eunuch assao snu sq usuaum “sash wuss-coo .:.m.: use ...-I sauna use .1 ca sundae: .m.4.n «can: ecu-asuasu .uuusvoum mud-a us .sounm ausueu you usesu / . . .. . . f..... H can “\.’.'.’..s s s o s s s s s s s s I s s sas - sesas '.-.... .r h o o s L“ -.‘.’ s. s. s o o s 0.0.0... ‘.-...‘.‘.-.'.-. ..A ’ r. . -.. I .‘.'.‘.. s .‘.n‘.‘.’.’.’. .‘.‘.‘ ‘ I Odd 038.732 .5322 .38 t.t.\\~ .w.= ..uusvoum undue no usuaum gasses new useeu .m.a.n can and novau owsus>< muqo .m.= you .m.a.a use scanned: .and-es; new gonna nose-coo .a.u.z assays» nauseous hue-e no escape sinusu I« concede any we assumed-so .u .nau 42 and 2. 6 to 13. 0 percent on the annual average basis. Over the eight- year period, the mean of the panel prices was 8. 3 percent below the B. L. S. mean. The difference between the two indexes was shown to be greater from the latter part of 1957 through December, 1958. The var- iation was mainly due to a slight decrease in prices of fresh milk and ice cream in the panel series relative to an increase in the same products in the B. L. S. series (Figure 2). The two indexes moved up and down together. Both B. L. S. and panel indexes remained rather stable during the first four years except for some intra-year fluctuations due to seasonal variations. A slightly upward trend during the last four years resulted from a small increase in prices of the items such as fresh milk, ice cream, and butter. Butter and ice cream prices followed a seasonal pattern which was similar to that of milk. The milk prices were lowest in the months of high milk production and highest in the months of low milk production. The seasonal pattern of this entire group apparently followed the price movements of milk, butter, and ice cream. These prices were lowest in early summer and rose to a seasonal peak during the winter. Both the panel index and the B. L. S. index of the fruits and vegetables group followed a pronounced seasonal pattern from a low in September and October to a peak in June and July, throughout the entire six-year period (Figure 3). The B. L. S. index was above the panel index in every 4-week period except in January, February II‘I|\I‘ I ‘II I -Il‘l' 43 anaa snag omaa Lan_________ ________a___________a_.2 :1 on .I oo . . cod . . . . lo: I s n \'.I.) u n .s \t’.’ u ‘4 . . . H..V tl..tt .1‘. .. ..\ s .m.‘ *.)..s o \ / e s \. s a I \ °N~ /.. .. .\ .f..\. . . .x . . \ .. \ 1o: . I .. r. ... . .\ I has . ( $2 23 _ .14. q a a .a _ _ _ _ _ _ m _ _ a _ a _ _ _ _ a _ _ m _ _ _ _ _ _ _ _ _ _ on ecuuum ss>qo onu ea souwum Hesse hols-coo .a.u.z vs. .s-sn souuh vs. enemas: .a.a.n .. I.om mew-D vouwu50u¢u ..s~n¢usus> use uuuShu we ssuqum Luau-s new aspen. .. or s I 0.. s . Sfl s o ‘ s s s ‘1. .s ‘I’ s .’u‘lu)n .x i .) I J . \.L..! I at.) . . .\ l. \1 .1 . . .1 cud - .. l ....\ s ‘.n' ‘.,a ..y .\ ...... x . . r. .. 833-53 .k I or .onsus>< huuo .n.= ..o~ns»sus> use sudshh we ssuuum gains: new “even .m.q.- cacao». auuo .m.= you .u.a.n use confined: .and-sea new Assam hula-sec .:.».x sessusa os~nsusus> was sudsuu we ssouua gasses su cousins OAu no se-«u-alou .n .nuh Noamfia 44 1957 and June, July 1958, over the six-year period. The mean of the panel average prices was 5. 9 percent below the B. L. S. mean. Comparing the annual average indexes, the B. L. S. series was from 3. 8 to 9. 7 points higher than the panel series, i. e. , the B. L. S. average price was from 3. 1 to 9. 3 percent higher than the panel average price (Figure 7). The two indexes conformed more closely at seasonal peak periods than at their troughs, i. e. , the panel price fell further when these products were in season. The B. L. 8. index fell off only moderately in season, mainly because the average prices were collected from 46 cities throughout the country, and these cities covered different climates and consumption patterns. The indexes, both B. L. S. and panel, for cereal and bakery products were comparatively stable during the four-year period except for the more irregular intra-year fluctuations in the panel series (Figure 4). These slightly irregular fluctuations were probably due to changes in qualities of some items which were pur- chased by the panel families from week to week. Because of the nature of the products in this group, neither series showed a pronounced seasonal variation. The two indexes did not follow each other very closely except that both had a slightly upward trend over the four-year period. The greater increase in the B. L. S. index was mainly due to the rising price of bread in the B. L. 8. series compared with the panel series. ’45 mama , Nam.— omoa , km or 4__L_L_aaafia~aqdann_L_L4_fi4__:.4fi4.amfi.a~jaaj verso.— sszu on» 5 :3: ..sssm hula-s8 .:.»... was .33 03.3 vs. 3:39: £5.- 935 6333310 .3263.— husxsu vss «ssh—so we use“: :38. yea asap . O s O s O s C o s . s s . O . . O s I o O s O s s s O o s a s s O o 0 s ‘1‘. .‘c 1.3. 9.-.- s s.- o ‘o.o‘o-.-o-s-o'.. .:- - 31.9.... guflaeusema .sushsbd 5:0 6.: .3023..— bsi. as. :88 no .82.. :33. new no.5 2....- . s s a O C. ... s o... 0o. s o’ss‘s‘s‘s’s s‘s-0,o 8.— IO: -.1 62 s .0 0 J I. and 1 03 emu...»- hzo .a... you 6...; was easier. 5333 new ~33 .:.-.38 .:.»... essays.— .uosvoua muons; use ..ssuso we :32. 33s.. 3 sous-do on» no sonausg .c .9: use: ' ...- ..‘ u .- “D. .u.‘ ... 1... "« ‘i‘ ..‘~‘ 0 "I N ‘A p ."v\ -> In ’A9 I. 1" 1.. 46 The B. L. S. index was higher than the panel index in every 4-week period. Over the four-year period, the mean of the panel average prices was 5. 9 percent below the B. L. S. mean. The B. L. S. index was from 2. 8 to 11. 2 points higher than the panel index. This in- dicates that the B. L. S. average price was from 2. 3 to 9. 2 percent higher than the panel average price. In comparing the annual average indexes, the B. L. 5. index ranged from . 9 to 13. 7 points higher than the panel index, i. e. , the B. L. S. annual average price was from . 7 to 11. 4 percent higher than the panel annual average price. From the appearance of Figure 5, it is quite clear that both the B. L. S. and the panel indexes in the "other food at home group" followed a similar pattern. The difference between the two indexes became greater during 1957 and 1958. This difference was due to a successive increase in the prices of eggs, sugar and sweets in the B. L. 8. series relative to the slightly lower prices of the same products in the panel series during this period. The seasonal price movement apparently resulted from the seasonal variation of egg prices. Both indexes were lowest in the spring and summer, during the months of high egg production, and highest in the winter, during the months of low egg production. Comparison of the two indexes indicates that the B. L. S. index varied from . 6 to 10. 9 points on a monthly basis, and 2. 9 to 9. 4 points on the annual average basis. According to the average price, the B. L. S. series was from . 5 to 10. 4 percent higher than the panel series on 47 unsus>s huge .a.» new .n.a.n was usauaout .and-aid new asset hols-coo .:.m.: sosauun slog us poem we‘ve no esuuua aususu ea eonssflo run we ao-uusaloo .m .uuh n3 R: o S 22 .__._.]:mml_.13_f___::-]__ __::mu; :_______.___.___ 3 I on I 8 I 2 I 8 vague» no>au enu an esouum «sash 5.528 5.9: can .28 8t.— ..5. 3.3!. 6:..- uSB ... co 32:53.5 ...-o: u. 38 “2.3 no 33.: 333. how 3.3/ . . . . /( cod 0 s.l 000's 0.. ...... 00¢ a O... ) .s 0.0 ...! \.\ ’2! I ) .o s .o \ ‘IIHP..hIo . o . ....\.a..0." )l.. . ... e\.\ II.I..I.¢.I.‘_ CA.— .I.‘. 1‘ I..’.I.(.\..\ ..‘.l...‘.‘ .I -.I.‘. 8~u§7§2 €933 58 .m... .Iom \ I o2 2. v8.— 5.8 no 83.: :3.- son .35 .....dn 1 as l 9: Novan !‘ a) p u- .:. h .... * - ”u r!- ’I 1 .- I I w- |.. I. ,__,_,_, ‘,_.__. . _ 48 the monthly basis, and Z. 6 to 9. 4 percent on the annual average basis. In comparing the two means, over the four-year period, the panel series was 4. 2 percent below the B. L. S. series. As indicated in Figure 6, the panel index for ”all food at home” moved almost parallel to the B. L. S. index. The price move- ments were rather stable and only a slightly upward trend was shown from January, 1955 to December, 1958. The B. L. S. index was above the panel index in every 4-week period over the four years. Among the time periods, the B. L. S. index was from '5 to 11. 6 points higher than the panel index on a monthly basis, and 6. 8 to 10. 3 points on the annual average. The indexes indicate that the B. L. S. average price was from 4. 9 to 10. 7 percent higher than the panel average price on the monthly basis, and 6. 5 to 9. 5 percent on the annual average basis (Figure 7). Over the four-year period, the mean of panel average prices was 7. 8 below the B. L. 5. mean. In summary, it appears that the changes in prices of the panel series relative to the prices in the B. L. S. series for the U. S. city average and Detroit (see Chapter IV) varied by groups of products. This is mainly due to the fact that the price change rate of each food item in the panel series was not exactly the same as it was in the B. L. S. series. If the price of each item changed at the same rate in both series, the two price series would move parallel to each other all the time. The comparison between the two series shows that the b9 mama » snag chad mama 44L:_w:_: d4—4fi_-__—_4—_4d_+11_~d_— vows...” ...->8 2: 3 :38 1a..— ..lsssu .:.»... on. .88 32.— can 3.3... 6:..- 3:5 32:93.6 ...-on a. 38 :4 no 83.: :33. ..8 ~33 / .fi-———d:—Adq°¢ 'l‘s‘iqs 8E..-SS .5825 58 .....s ..som u. econ -<.uo .aoauu nausea new news» .m.s.n ... can .632 llama ensue». 530 6.9 new 65.: was 93.33! 33.3 ueu uses.— sols-sou .:.a.: census: use: an veeu use no ssouua gasses ea sens-no sAu we seeuuuAIOO .o .3.— uses" . 'ICI.KIIII .IAUQU ‘ nu‘i I d yd-p-Iu -sdd .lfli‘ lit ... 's‘x‘luss ‘ -11].-.- Ivsl' ts‘-'fii HI." I l'l Is-t' ‘IlfiunIa- 'Iallll.a‘ I.uvI SO an pm on mm on an on an an an on no L _ _ _ _ _ . ca cog . . --. -----. . ..... “Knurfip -....... I o: 0“ s ......... s ....... s ......... J ONd \ \ \ \. - .l 2: lo: us poo.— Z< «In: as poo.— usnuo 303.3.— 333 v:- :88 on an on an «m an wn «n on an «n ma an an an hm on mm «m an «m an _ _ _ _ _ _ _ a _ _ _ a .4 _ _ _ 4 cm .. . 1 8 s. 0. \sII . . s u ‘ [-4 O \\ "0 8d . \ I .. O ..... s. .I ‘0“- n .- s\ [I .- . e. \\s\\ O/I/sl . a \\\ \\ loll,- \\\\sll’ \\-\ IIe\ \ I/ 1 0d.— o . . o \o I. o\ N I .. .---» \\ ...: ..---. .a.\\ . nuosoehm hug-a was .huufisom .oussx .1 ON“ .. x 3.330»; was .33.: \ . seven 833-32 ...—32:. huuu .m.= .sa56um poem vsuwaooas «an new nexsvsu .a.a.n we soueuo>s Anseluuu Illll' weaken so>qu 059 :« esouua asset hols-sou .:.».r pas .sesn sauna was euauugs .n.4.n Isstusn sasouu 18v No .00.?!— nusasu flu Iona-£0 33 NO consul». discus Ola no gsuh'llou .u anus: usulasuaso .nasouu 100w vsauaoea- one you ssxovsa on» we asusns>s nausea-n~ 9:2.; onsu0>s huuo .m.= you .m.4.n pas sinusoit .uausssa new asset ails-sou .:.».x . INK ‘0 ‘OD 51 B. L. S. quoted prices from the specified products were usually higher than the panel prices. This is mainly due to the fact that the B. L. S. specified qualities were usually above those qualities actually pur- chased by the bulk of consumers. Moreover, some of the panel families might buy some cheaper food items in the farmers' market or other sources. ‘Nevertheless, the parallel price movement provides quite satisfactory results in comparing the two series, even though there are some differences of qualities between the two samples. The Relationships between the Two Price Series It has been noted in the preceding paragraphs that the price series of the B. L. S. and the M. S. U. Consumer Panel moved up and down together over time. However, the differences between the two series are widely divergent as shown in the accompanying graphs. The variations between the two series may not be exactly correlated with each other if they are based on each observation from one period to another, even though they have similar patterns of price movement. Thus, it would be of practical value to measure the interrelationship between two price series as measured by the coefficient of correlation, i. e. , to measure of the closeness between two series in which the prices were originally expressed, either by an individual product or an average of a group of products. The correlation and regression coefficients between the two series usinLoriginal prices. --The actual prices of the two groups, 52 'tereal and bakery products, " and ”meats, " were used. The reasons for choosing these two groups is that the prices of cereal and bakery products did not follow the B. L. S. movement very closely, while meats followed each other very closely, as shown in the graphs. These two groups of products represented the extremes, and, therefore, were selected for correction analysis. The estimating equation was: y = a + bx where x 2 prices of B. L. S. products for the U. S. city average y = prices of the M. S. U. Consumer Panel products A series of simple correlation coefficients "r" and re- gression coefficients "b" were calculated along with the "tb values, " and standard error of correlation coefficients " {1", " which are summarized in the following tables. It has been shown that the two price series, for cereal and bakery products, did not move very. closely with each other (Figure 4). Because of the nature of the products in this group, the correlations between the two series were not as highly significant as they were in other groups. As they are shown in Table 1, only four of the nine products were found to be significantly correlated with each other in price movement between the two series. When dealing with the meats, there is obviously a high Correlation between the two series, for each product in the group (Table 2). Their correlation coefficients are consistent with what is .‘s‘ .. 53 TABLE 1. -~Summary of regression and correlation results for cereal and bakery products, using B. L. S. prices for U. S. city average and M. S. U. Consumer Panel prices, 1955 to 1958 Level of Products r of b t value significance (1‘)** (tb)** Rice -.08 .14 .68 1.42 n.s.* n.s.*' Rolled oats . 6’7 . ll . 71 2. 61 1% 5% Corn flakes . 85 . 08 1. 39 3.14 1 l Flour, wheat .14 . 14 -. O6 16. 87 n. 1 Biscuit mix -.01 .14 .05 1.53 n. ma. Corn meal . O4 . l4 -. 08 3. 88 n. 1 Bread .4? .12 .22 13.13 1 1 Soda crackers . 52 .12 .30 10. 00 l 1 Vanilla cookies . 03 .14 . 09 2. 14 n. s. 5 .9. Not significant at 5% level with 50 degrees of freedom. #31! George W. Snedecor, Statistical Methods (Iowa State College Press: Ames, Iowa, 4th ed. , 1946), Table 3.8, value of t, page 65, and Table 7. 3, test of null hypothesis, P = 0, page 149. The significance of the regression coefficient is tested by the calculation of t from b,-l i 1 lb. 1 where ti is the t value of ith regression coefficient being tested sb is the standard error of bi. ‘ i shown in their price patterns as they move up anddown together over time (Figure l). The regression coefficients indicate that the changes in magnitudes in the prices of B. L. S. series were associated with the changes in the prices of the panel series. When applying regression analysis to market statistics for the purpose of making estimates of 54 TABLE 2. --Summary of regression and correlation results for meats, using B. L. S. prices for U. S. city average and M. S. U. Consumer Panel prices, 1955 to 1958 Level of Products* r of b tb value significance (r)-"k* (b)** Round steak . 95 . 03 . 89 3. 54 1% 1% Chuck roast . 93 . 04 . 88 5. 99 l 1 Rib roast .91 .04 1.11 2.11 1 5 Hamburger .95 .03 .97 1.01 1 ms. Veal cutlet . 71 . 07 . 63 5. 84 1 1 Pork chops . 83 . 06 . 88 7. 07 l 1 Bacon . 96 . 03 . 86 5. 78 1 1 Ham . 76 .07 .89 4.81 1 1 Frankfurters .95 .03 .93 2.17 1 5 Lamb, leg .47 .09 .82 1.13 1 ms. Lunch meat . 92 . 04 1. 30 4. 74 l l :9. With 99 degrees of freedom. ** Snedecor, op. cit. TABLE 3. --Comparison between the mean of Panel prices and the mean of B. L. S. prices for U. S. city averages** Products* B. L. S. Panel Products B. L. S. Panel Rice 17. 67 26. 54 Corn meal 12. 78 ll. 72 Rolled oats 20.18 15. 21 Bread 18. 46 18.12 Corn flakes 23.18 31. O6 Soda crackers 28. 20 2.8. 71 Flour, wheat 10. 80 9. 34 Vanilla cookies 55. 43 43. 38 Biscuit mix 26.92 17.76 Round steak 97. 11 80. 22(1) Chuck roast 57. 98 63. 09 Ham 65. 10 62. 67 Rib roast 75. so 6109(2) Frankfurter 58. 37 55.50 Hamburger 47. 50 47. 64 Bacon 70. 86 58. 37 Veal cutlet 118. 28 74. 02 Lamb, leg 73. 86 68. 81 Pork chops 82. 83 73. 78 Lunch meat 46. 67 57. 30 zit The eight—year mean of meat prices, and the four-year mean of the prices of cereal and bakery products. *1? M. S. U. Consumer Panel and the B. L. S. Monthly Labor Review. 1All steak. 2 All roast. 55 two price changes, the question arises whether or not the variables dealt with should be removed of their trends before the regression analysis is commenced. We note that if a trend is present in the price movement this may dominate the short-term fluctuations. Trends may give rise to spurious regression. However, if the trends were removed before the analysis, this would throw away some of the statistical information that is available on the relationships studied. Trends are the long run variation of the variable considered, whereas the short-term variations are the frequencies of. fluctuations. The short-term fluctuations include cycles, seasonals, and irregular variations. Hence, if we work on trend-removed data, the resulting regression coefficients should be regarded rather as reflecting short- term changes. With regard to the regression coefficients based on data with trends included, these appear to reflect something inter- mediate between short-term and long-term fluctuations. Since our primary interest was to compare the amplitude of price fluctuations between the B. L. S. and panel series rather than the frequencies of price changes from period to period, the trends were not removed before estimating the price changes of the two series. If there is a trend, the effect of the trend tends to bias the regression coefficients "b" toward 1. The greater frequencies of fluctuations will bias "b" downward. The regression coefficients may either be biased by the trends or by the frequency of fluctuations. As shown in Figures 1 through 7 (also Figures 8-14 in ..l. .. . .as o... Ii. His ‘1. .I- .11.. ..h.‘ ....N «In \I- s. 10 56 Chapter IV), there was not any food group showing a clear trend except the ”cereal and bakery products“ group which showed a slightly upward trend in the B. L. S. 8612.65. It has been noted that there were greater frequencies of fluctuation in the panel series. Therefore, the re— gression coefficients may be biased by the greater frequency of fluctuations in the panel. The smaller ”b" coefficient in the ”cereal and bakery products“ group was possibly biased by the slightly upward trend in B. L. 5. series and greater frequency of fluctuations in the panel series. The least squares regression of the panel series on the B. L. S. series (for the U. S. city average), using the actual prices, produced a series of regression coefficients "b" (see Tables 1 and 2). In the tests of the regression coefficient of each food item in the "cereal and bakery products" group, all the items in this group were significantly different from 1 except rice and biscuit mix, which were not signifi- cantly different from 1 at the 5 percent level. Among these items, only the prices of corn flakes in the panel changed wider than in the B. L. S. series. The negative regression coefficients show that the prices of wheat flour and corn meal between the B. L. S. and panel series changed in the opposite directions. This indicated that the panel prices went up while the B. L. S. prices went down and the-panel prices went down while the B. L. S. prices went up. This was due possibly to the effect of the greater frequency of the price fluctuations in the panel series. The B. L. S. prices increased while the panel a PU a. ‘9‘ I“ .... . I.; -.. ‘0‘ H I n‘ :‘l \_‘ '4 1". u u w, E. “.‘ 57 prices fluctuated up and down from one period to another. When dealing with-the "meats" group, the regression coefficients of the items were all significantly different from 1 except hamburger and lamb. Among these meat items, only the prices of rib roast and lunch meat in the panel series fluctuated wider than in the B. L. S. series. The panel prices for other meat items fluctuated less than B. L. 5. series. (See Table 2.) In comparing the average prices between the two series, the means of the B. L. S. prices are not all higher than the panel means. The high prices of rice in the panel were due possibly to errors in reporting of purchases by the panel families. The higher panel for corn flakes were due mainly to the fact that the B. L. 5. series on corn flakes was based on the prices of the large size packages but the purchases of corn flakes by the panel families were in both large and small packages. The panel purchases were not adjusted to exclude the effect of the greater cost per ounce of corn flakes in small packages. However, most of the B. L. S. products which are highly weighted in the index have higher means of prices than those in the panel series. Since the price index is computed according to the weight of each product in the group, the B. L. S. indexes are always higher than the panel series (Figures 1-7}. The correlation and regression coefficients between the two series using their index.--The correlation coefficients were cal- culated by using indexes for each group between the two series. The 58 estimating equaticn was: y = a + bx where x : index of B. L. 5. groups of products for U. 5. city average, y = index calculated by using B.L. S. weights and price base but M. S. U. Consumer Panel prices in the given periods. The results of the regression analysis are summarized in Table 4. TABLE 4. --Summary of regression and correlation results for groups of products, using the index for U. S. city average and the index calculated with the B. L. S. weights and base, and M. S. U. Consumer Panel prices in the given periods Level of Products"? r a? b tb value significance Meats, poultry and fish . 95 . 03 . 95 1. 57 1% n. 8. Dairy products . 43 . 09 . 50 4. 76 1 1% Fruits and vegetables . 91 . 05 l. 25 3. 89 l l Cereal 8: bakerypr oducts . 48 .12 . 24 12. 37 1 1 Other food at home . 48 . 12 . 74' 1,. 36 1 n. 5. All food at home . 91 . O6 . 66 7. 71 l l a Meats, poultry and fish and dairy products are tested with 99 degrees of freedom, fruits and vegetables with 76 degrees of freedom, and the other groups all with 50 degrees of freedom. my Snedecor, op. cit. 59 The correlation coefficients between the two price series, as measured by the index of each group, are all highly significant. When measuring with the original prices of cereal and bakery products, the correlation coefficients are not all significant in terms of each individual product (Table l). The high correlation coefficients, as measured by using the indexes, are mainly due to the significant products with higher weights in both indexes such as bread and soda crackers. The apparent relationship between B. L. S. series and panel series can be explained by the results of regression analysis as well as the graphs. Obviously, the groups of products move closely with each other between the two series when their correlation coefficients are highly significant. The least squares regression of the panel series on the B. L. S. series (for the U. S. city average), using the index of each food group, produced a series of regression coefficients “b, " as shown in Table 4. The tests of regression coefficients of each food group indicate that the ”dairy products" group, ”fruits and vegetables” group, ”cereal and bakery products'l group, and "all food at home" group were significantly different from 1 at the 1 percent level. The ”meats, poultry and fish" group and the "other food at home” group were not significantly different from 1 at the 5 percent level. Among these food groups, the prices of the "fruits and vegetables" group in the panel series changed more widely than in the B. L. S. series. The , h r s ... .l - . ... ... n .n , ... u s , s a .1. :~ ‘11.. hi! t 60 prices of other groups changed less in the panel series than in the B. L. S. series. As shown in Figure 3, the panel prices of the "fruits and vegetables" group changed more widely than the B. L. S. prices. The panel prices of the ”fruits and vegetables" group went up further when they were out-of-season, and fell further when they were in season. The "b" coefficient of this group calculated by using the indexes of the panel series and the B. L. 5. series was 1. 25. This indicated that a 1 point change in the prices of ”fruits and vegetables” group in the B. L. S. series was associated with a change in the same direction of l. 25 points in the prices of "fruits and vegetables" group in the panel series. This result was consistently related to the comparison of the price movements between the B. L. S. and panel series as shown in the graph. (See the analysis of Figure 3.) The regression coefficient of the dairy products group was . 50. This indicated that a 1 point change in the prices of "'dairy products" group in the B. L. S. series was associated with a change in the same direction of . 5 points in the panel series. As shown in Figure 2, the amplitudes of the price changes in the B. L. S. series were greater than the panel series. The panel prices of the ”dairy products” group changed more frequently from period to period. This resulted mainly from the qualities of the dairy products purchased by the panel families from one period to another. When dealing with the "cereal and bakery products" group, I ... n. as p.’ 'e 2.! a” ll" ed 61 the regression coefficient was . 24, which was significantly different from 1 at the 5 percent level. This indicated that a 1 point change in the "cereal and bakery products" group in the B. L. S. series was associated with a change in the same direction of . 24 points in prices of I‘cereal and bakery products" group in the panel series. As shown in Figure 4. the panel prices of ”cereal and bakery products" group showed little changes over the four-year period except some short- term irregular variations. However, the B. L. S. prices of the "cereal. and bakery products” group showed a rather clear upward trend from 1955 to 1958. The irregular variations in the panel series were probably due to the changes in qualities of the food items in this group which were purchased by the panel families from week to week. For the comparisons of "cereal and bakery products” groups between the B. L.S. and panel series, see the analysis of Figure 4. The regression coefficient, calculated by using the indexes of "all food at home“ groups of the B. L. S. series and the panel series, was . 66. ‘ This indicated that a 1 point change in the prices of ”all food at-home” group in the B. L. S. series was associated with a change in the same direction of . 66 points in the prices of "all food at home" group in the panel series. As shown in Figure 6, the B. L. S. prices of the ”all food at home” group changed with greater amplitudes than the panel series. This was due primarily to the fact that the panel had smaller amplitudes of the price changes than the B. L. S. series in each food group except the ”fruits and vegetables” group. The wider 'p; I . A -- 0,. LJ I" :1- Q . ‘u i n‘ . ‘5 H. ‘v. 5‘. 62 changes of the "fruits and vegetables" group in the panel less than offset the smaller changes of prices in each of other food groups. When dealing with the ”meats, poultry and fish“ group, the regression coefficient was . 95. This indicated that a 1 point change in the prices of "meats, poultry and fish" group in the B. L. S. series was associated with a change in the same direction of . 95 points in the prices of the ”meats, poultry and fish" group in the panel series. As shown in Figure l, the price changes of the "meats, poultry and fish" group in the B. L. S. and panel series were about the same. If the changes of the two price series were in the same proportion, the regression coefficient would be equal to l. The regression coefficient would not be significantly different from 1 at any level. In this case, if we know the price changes in the panel we also know the price changes in the B. L. S. series. This relationship can be applied to each food group as well as to each food item. . As shown in Figure 5, the‘prices of the ”other food at home" group in the panel series and the B. L. S. series moved up and down together very closely. However, there were no appreciable differences of the price fluctuations between the two price series, as indicated by the movements of the two indexes in the’graph. This relationship between the panel and B. L. S. series (for U. 5. city average) produced a regression coefficient . 74, which was not significantly different from 1. The regression coefficient indicated that a 1 point change in the prices of ”other food at home” group in the B. L. S. series was 63 associated with a change in the same direction of . 74 point in the prices of "other food at home” group in the panel series. nu .- ._ ~\~ .tw 64 CHAPTER IV COMPARISONS BETWEEN THE M. S. U. CONSUMER PANEL SERIES AND THE B. L. S. SERIES FOR DETROIT The Changes in Prices between the Two Series Because the indexes, one based on B. L. S. data and the other on panel data, are calculated by using different price bases, a comparison shows only how much prices have changed in Detroit compared with Lansing from one period to another. After the B. L. S. indexes are converted from 1947-49 = 100 to 1955-57 = 100, com- parisons do not show absolute differences in price levels, since the average 1955-57 prices in Detroit were not exactly the same as in Lansing. Nevertheless, comparisons of the two indexes can be facilitated by this conversion since they are measured from the same period of time as a base. An examination of Figures 8 to 14 and 1 to 7, respectively, shows that the B. L. S. index of each group of products for Detroit appears very similar to its index for the U. S. city average. It is also noted that the panel index calculated by using B. L. S. weights for Detroit (Figures 8 to 14) appears very similar to the panel index calculated by using B. L. S. weights for U. S. city average (Figures 1 to 7). The changes of the two price series between the B. L. S. for Detroit and the panel for Lansing are the same as the B. L. S. for the U. 8. city average and panel for Lansing, which have already been 65 TV? presented in Chapter 1...... The following figures, 8 to 14, are the graphic presentations of each group of products between the two series. In addition to the comparisons of the price movements which have already been stated in Chapter III, a brief analysis of each group, accompanying its graphic presentation, is concerned with the main differences between the two series. Apparently, the long-time trends of the B. L. S. index and the panel index in the "meats, poultry and fish'I group followed each other fairly closely. Both series followed a downward trend in the first four years and an upward trend in the last four years. However, the price fluctuation in the panel series was less in degree than the B. L. S. series for Detroit. Over the first four years, the panel price was relatively stable. Figure 8 indicates that the B. L. 5. series for Detroit experienced relatively greater price declines during the period from the latter part of 1952 to the end of 1954. This mainly reflected the fact that the meat price in the B. L. S. series for Detroit dropped further when meat prices were decreasing and went up only moderately when they were increasing. In comparing the annual average prices, the B. L. S. series for Detroit was 17. 4 percent higher in 1953 and 9. 0 percent higher in 1954 than the 1955-57 average, while the panel series was 13. 5 percent higher in 1953 and 12. 5 percent higher in 1954 than the 1955-5? average. The annual average prices in other years were about the same in both series (Figure 14). It should be noted I) I'lurluu- III‘} .‘III‘IIII- i 'll \ 66 onau snag onau “nod _ __ _ _ _ __ _ — __ — _ __ _ _ _ —_ _ fl _ _ _ L __ _ _ _ ._ _ _ _ __ _ _ _ _ q a 05 I; on ....»..»..u...+. j I 8 ..... .\ .....I: . \... .... ./ LEV,“ I I, 0.).rl-‘J . 0., ul ban, a .- ..«...\ar.../ .0....‘ ....N...r...u.$.. 0 COM ..\.... ......I ...“ ...u... ....\..\ . v3.5. 350 J o: .;;.u.r. .... .uau nu acouum don-m hula-coo .9. m. x can .OQuusno nnaa oouwum lunuo>< mos-m hols-coo .:. w. x .uuouuoa haw Keven noon flanges nu sundae: osae> «0 oonuquAIu o>«unaqu .m. A. n I: cud wean: veunanoauo .:.«h vs. .huu~:om ..unon no noo«um u«eaeu new noun” :1 and «nag nmo~ «nag “nag ___q_______d _~__aq_________«_fi__~___~__~________1om :1 cm ... b.) cod .. ..u.’...l ‘20 \- I \ \1 D I o. I. l..'....|. .J‘W‘... .... . ..\. ; .. a .u x . ...\ . .-....r: l o: .o 1“. o s 4.” n n. I .o. o .... 8E“... -22 3 3:268 ., .........\.- .-...-........n.... ..---. .SWS- $2 .3938 .:.: 2:. 5:2: \. k .«2 L. .-.} I O2 .33: no :3: 33:. new .82: .m. ... .— «gunnnnmwflgnu nauuzouz .uuouuon new .m.a.n vac auaunuux .uuuna-A new denim nonunuoo .:.u.x cootuon scum vac .huuusoa .eu-II no noouua guinea aw nemcnnu onu mo ao-uucnlau .m .uuh 67 unoa an wm%u wl_JI._I._.JI_|_.I_ITT~—________m_...:a4qfi:___ :___—:___m_m.flqq__q 2 I. on I. ca . . .....HHU . . . ...... . .... .....I a I.-.I.....I.I..+.I; « ......y I-.-.I-.-l1.rp.......fluff 1-4.... . .. 2: I. odd I. and I. and «nag nnaa «nod anou a fia_____:_:_ _:a____q+fi4 __________a__::qfi_qfiafia o . . I. oo Ocansnunnoa 0» vouHODuoo Ocauotuncou vacuuon !I:..I.I.I.I.p./ .cuoavoum huuda no noowum fluent: wow Kevan .m.4.n//////K . ............. 11%.f...(.' 1.'.‘.I.‘.‘..I.l.'.'.f. ’ O \.l.\..0 1.....‘.Av.-.|..\.x O0." . lo: 8:: .330 of 5 33.: 7.5— ulauoo .:.»... vs. .8732 Inmma eouuum owuuo>< «ea-m hols-coo .:.».x .uuouuon new lovau I. o- voou au-uou ad canned: osan> we condense-u o>qu¢acu .m.A.- ”can: veg-aauduo ..uoavoum aka-a no nuuaum “anus: new Novau Agdlnansvuoufl aluunuuz .uuouuon new .m.q.n van undiscis .and-cud new ~ealm boss-:00 .D.u.: :ocluon euuavoua hue-v no cough; dude-u nu noun-no enu no needy-aloe .o .uqh 68 that despite the wider fluctuations of Detroit prices, the B. L. S. series for the U. S. city average (Figure l) fluctuated less in degree than both the panel series and the B. L. S. series for Detroit. This moderate fluctuation resulted chiefly because Detroit was averaged in with 45 other cities. It is evident from the foregoing presentation (Figures l-7) that dairy product prices are more stable than the prices of most other agricultural products. Figure 9 shows that the eight-year trend of both the B. L. 5. series and the panel series in dairy products has been stable and that they followed each other fairly closely. During the period 1951-1954, the panel index moved above the B. L. S. index for Detroit. This difference is mainly due to the fact that the 1955-57 average price of the B. L. S. series for Detroit was higher than the panel 1955-57 average price for Lansing. After the conversion, the B. L. S. indexes became lower when their base prices were higher. From the latter part of 1957 to the end of 1958, the B. L. S. index for Detroit moved above the panel series. This corresponded with the same period in Figure 2., in which the B. L. S. series for the U. S. city average diverged widely from the panel series. On the annual average, the B. L. S. index varied from a low of 99. 3 to a high - of 103. 3, compared with the panel series which varied from 101. 2 to 107. 7 in the first four-year period; and then in the last four years the B. L. 5. index varied from 96. 6 to 102. 2, while the panel series varied from 96. 6 to 101. 7 (Figure 14). 69 It is evident from Figures 9 and 14 that from the latter part of 1957 the panel index declined slightly as the B. L. S. index continued to rise. This was the same result as was indicated in Figure 2, i. e. , that there was a slight decrease in prices of fresh milk and ice cream in the panel series relative to an increase in prices of the same products in the B. L. S. series for Detroit. The indexes of the B. L. S. and panel series in the fruit and vegetable group moved closely with each other during both the seasonal peaks and troughs. As shown in Figure 10, the panel series rose further when fresh fruits and vegetables were out of season and fell further when they were in season. That is, the B. L. S. series for Detroit fluctaated less in degree than the panel series. As shown in Figures 10 and 3, it is evident that the four indexes moved very closely with each other. It has been noted in Figure 3 that the mean of panel average prices was 5. 9 percent below the B. L. S. mean of average prices for the U. S. Since the B. L. S. prices for Detroit were about the same level as for the U. S. , the difference between Figures 10 and 3 occurred chiefly because the B. L. 5. 1955-57 average price for Detroit was higher than the panel average price in the same period. - On the annual average, the B. L. S. index varied from a low M91. 4 to a high of 102. 3, compared with the panel movement from 92. 3 to 103. 7. As shown in Figure 14, the B.L.S. annual average indexes rose steadily from 1953 to 1958, and the panel series had a Blight decline in price during 1957, after three years of successive 70 increases, and then went up again in 1958. The prices of cereal and bakery products remained relatively stable largely as a result of two circumstances. In general, (1) the products in this group were less seasonal in nature, and (2) their demand and supply were relatively stable. As shown in Figure 11, the panel index experienced greater intra-year fluctuations than the B. L. S. index. These fluctuations were mainly due to the fact that panel families purchased di f‘erent varieties of products from week to week. The B. L. S. series moved above the panel series from the latter part of 1957. This difference corresponded with the same period in Figure 4, in which the B. L. 5. series for the U. S. city average showed an increasing divergence from the panel series. On the annual average, the B. L. S. indexes varied from a low of 98. 8 to a high of 103. 7 compared with a variation from 99. 4 to 101. l in the panel series. As shown in Figure 14, the panel annual average prices of 1957 and 1958 were about the same as the average prices of 1955-57. However, the B. L. S. annual average prices for Detroit in 1957 and 1958 were about three percent above the 1955-57 average price. The price movements of both the B. L. S. and the panel series "other food at home” group are shown in Figure 12. If this chart is compared with Figure 5, it will (be observed that the four different indexes moved very closely with each other. From the latter part of ' 195?, the panel index declined more sharply than the B. L. 5. index for Detroit. This was the same result as was seen in Figure 5, chiefly D Inuit. II'.uIOI|l 71 anon snag one“ 4____d_d.___ .___._______ ________—q—‘°h I.ow . .. ._ .. . 1 2 f .n 9 ...J. “1 4| . ).... \r .\r .” snek u.( w/ Amt .l .n.( .l k n u 9.\ I\ > N .. .u I a. \ .. . v r .s 4 \ . 8.. r... V ..\ ’V....\ . ... . x ./ A A... l o: ... ....01. 1.. w. . - I0-o *1“: J o2flTnns o2 cu vauuo>aoo .oo~laetnca~ .uuouuon ..oaneuomu> use onusuh we oeuuum fiauucu you xoveu .m.a.u 102 22 $2 / 22 _______4_._ ___________/____.H.......,________8 . he}... .32.... ‘..)n h. I‘ \..l/ a ».1, a; .9 . .:.1: \. .9 . o ..l‘o t, o\ .- .. .. loo .. 0‘ .- ‘ I .. / t\.\.t.\ / «2.36.. - ../ NVK d. . . .iuhlep...lon o\ . I a... d ‘5. all. .. .» ..\ .. . x... a. ...... ..s f . . (.... A. v. . . ... .... cu. CC lieua ecuuom ao>uu any a« noouuu gonna unis-coo .:.m.: vac .oounsnonmmu acouum ounuo>< Hon-m hols-coo .:.m.x .uuouuen new |.o~a gonna voou gauged a“ nunuuos osan> no ounduuoalu «>«uuaeu .n.a.n wean: veg-asod-o ..oda-uowo> vac unusuh we unsung auqueu uou novaa AOOaflsnannoavnuvnH auaunout .aaOuuoo new .m.a.n can cqaunuux .and-=14 you mecca unis-sou .:.a.: sooty-n coda-aunv> use nausuu no noouua aquuou ca nous-no 0:» no cough-aloe .o~ .uuh 72 anon and 33 an: fl__________~.——___du___________—__u-__——._-__——._._- 3 I. on I on 1 an I an Suflnnnns cu v3.5.5.3 6333-32 .3038 .3269: Dead- alcuoo us .03.: .3308 new Kevan .w..—.n / .1 co .. .. .. 9.........a.»..t....«tu.r..q...w....a..a.i..........e...t...tw.t.un..n...-.-.q 2: o. ..-. -.h-.-.'n‘r -.-...P...h..'...-.o..h‘.‘....-9.3.4".-..“..‘o‘. .... ...... I. o...— vouuom .530 on» 5 :0un ~38 Malia—8 .:.u.‘ can 63.45932 :3: each—£5 and: hula-coo .96.: .3958 new nova.— voom .338— 5 35:03 03o> «0 00.30.33 0233.: 6:... l 0%— uS-p 133538 .3033; hues- vs. «sou-o we .03.:— :3!- uou .313 I 03 . l 9: ASdlntnnacusau conic:- .u«ouu8 new 6.3.- 1.. .333: .333 new no.8 nag-:8 .:.ni in.» 3038...— :33 vs. :33 no .03.:— 333 n.— ooualo .5 no sour—lino :3 6: 73 owed Anna onma nnau cc _ _ . ._ a a __ _ _ _ __ . _ _ a an _ _ _ q _ _ a _ ._ _ _ _ __ _ _ _ _ _ _ _ _ _ 1a a a __ |.om I.oo I.o~ [mow ooaflsntnnaa 0» vouuo>aou .ooHnfieuucmd .uaouuoa .olo: u- voom nosuo no nonuum ”aquou you nequ .m.a.n ca 0 ..N. ... o o o o In. ... a a .... ..o. ‘...’...vt.‘ .I. V.).’ l. ....‘ ... ... £....\....)...LH \fi...n \1 ......r‘c‘flr. ....nr. .....9. ....L n} .1 A, 4‘ , r‘fl-Ufi-IJdfllth‘ r’ .‘.-u..'...,u_.o... . ....J‘...(.‘ V .H...I...‘...‘.’b‘.r1p" .( t’ 8H I.od~ vauuom ae>uo any ad conqum doe-m hols-:00 .:.m.: use .oeaNsntnnoa nouuum oueuo>< “seem unis-coo .:.m.: .uuouuoa new arena soon adders ea eunuaoa oaau> no sou-unease erased-u .u.g.n |.o- mew-D tau-Asuaqo .ulo: an voom wonuo we noouum away-m new woven I.cn~ m 9: Sofienéacuova aquusoux .uuouuon you .a.A.- can ndaunoux .and-neg you don-m hols-ecu .:.m.x sooty-A «Ion us coca weave mo noouun uueuou ea conunnu oAu no :o-«ucalco .NH .uan 74 because the panel prices in eggs, and sugar and sweets were lower than the B. L. S. series of the same products. On the annual average, the B. L. S. index varied from a low of 98. 7 to a high of 100. 6, compared with a variation from 95. 5 to 100. 5 in the panel series. Figure 13 indicates that the panel index in the ”all food at home“ group fell much lower than the B. L. S. series from 1957 to 1958. This difference resulted mainly from the fact that the prices of most of the products in the panel series increased at. a relatively small rate in comparison with the B. L. S. series. The panel index moved above the B. L. S. index from the beginning of 1955 to the middle part of 1956 and then dropped below it from the latter part of 1956 to the end of 1958 (Figure 13). In comparing Figures 13 and 6, it is evident that the lower prices of 1957 and 1958 in the panel series corresponded with the higher prices of 1957 and 1958 in the B. L. S. series for Detroit as well as for the U. S. city average. An examination of the B. L. S. index for Detroit and the index for the U. S. city average indicates that these two indexes were about the same throughout the entire period. Since the B. L. S. prices were higher than the panel prices (Figures 1-7) and the B. L. S. prices for Detroit were about the same as for the U. S. , it is quite obvious that the B. L. 5.1 prices for Detroit were also higher than the panel series. As shown in Figure 13, the lower level of the B. L. S. series was mainly the result of itshigher price base compared (with the panel series. On the annual average, the B. L. S. index varied from a 75 low of 96. 9 to a high of 103. 4 compared with a variation from 97. 6 to 100. 2 in the panel series. In summary, comparisons between the panel series for Lansing and the B. L. S. series for Detroit were very similar to the comparisons between the panel series for Lansing and the B. L. S. series for the U. S. city average. The magnitude of the price fluctuations might have been somewhat different, but the patterns of movement were about the same. The conversion of the index to the 1955-57 base only shifts the level of the index. Nevertheless, the percentage change from one period to another remained the same. The Relationships between the Two Price Series The correlation and regression coefficients between the two series usingorigi‘nal prices. --Only cereal and bakery products and meats were calculated by using original prices between the two series. The estimating equation was: y = a + bx where x = prices of B. L. 5. products for Detroit and y = prices of M. S. U. Consumer Panel products. A series of simple correlation coefficients ”r" were cal- culated along with their regression coefficients "b, " "tb values, " and Starldard error of correlation coefficients " (f, " which are summarized in the following tables. The correlation coefficients in Tables 5 and 6 76 :3 Ana.— onfl nnfl Aaqq___a__a4~__an_44__n_. .fia. ...«qqada.‘__q4a.q..+713 l on 1 8 L 2 l on 0319.33 3 33258 .8Huactsz .3933 .olo: u. voou :4 «a goof: .33.: new N33 533/ 1 ca ............. .:.... ..-.....i titttct» .... strutttcotttun .3... ..l .1... -.....t:......-. ..t ...... ..l. as ......F...»........Ht. \ .\ \ v0.3..— 6050 any 3 .00.“: d2...— 1 o: acne-e8 .:.m.: use .8~..knunn$ :3: 3322 gonna hula-sou .:.»... .3933 new :3 voou :31- 5 343.3 .3; no ouaduuoaln Sauna-u .mJ.‘ new-a wound—.38 .ulom u. vooh :4 we :3: :3!- uou nova“ 102 long 10.: «gangs 93.33: .3938 new $5.- 33 damaged. .9333 now need.— unis-a8 .:.»... 3.252— quA u. .33 :- uo .03.:— 33: 3 .09—30 .5 we annual—loo .3 i: n on 3 on n... on R 3 an on R on 2 .41\ _ _ _ _ an _ .a. co ... \.‘|.|||.I|A . | .|| .\.-. ..\..\.v.1.............:. lllll .Hu.lll|n.f.. |\\\ ........... - \\V \~I| I .... cod 1 o: I .5 elem u. vcou Ha< clam u: soon nonuo auoavoum huoxcn can unouoo an ona an mm on an Qn an an so on an «m an «m an mm mm on an «m an an an d _ _ d . . _ . . _ 4 . d . . \1 .q on ‘0 "l; J om fin“... ........ ‘ \.\.\ . \..\.u.H.I|-II xvii. ..z. dR: e:.1!. Wm. .:a Ir. & >\ J! "nu-‘V‘C .0 o|||o\l‘au] ...... Il’0\\ \HKfi .JV. 8H ...-.:.. a N... lull .\. ...I/ II a. ....V . o: ceanuuouo> van nausun guuom HIN—u «nu a— Human“ #0:: u!- D m z hula-coo .:.».l .uuouuon now novau coca Hannah nu ousuuo: Gaga? no oonluuoalu o>uu¢aou .m.a.n undo: voucaaonuo .ansouu ween vouuuooa- you couovuu on» no .ounuo>u vo«uoa-n~.:..: Saflsnunnou cu wouuo>u00 IIII .oo~ns<-h¢oa .uuouuon .oasouu wean cognac-A. you aux-flaw .m.a.- mo non-no». auaqlona unsunUul .uuouuon Maw .m.q.- wad niaunuut .and-and you denim hula-ace .:.m.z Iuoiuoa auscuu vOOu no noowun uqduoh nu Iona-£0 Gnu uo IONIMO)! ulflflnc can No aeouunalao .Qu .nuh 78 show that the relationship between the panel prices for Lansing and the B. L. 5. prices for Detroit was about the same as the relationship between the panel prices for Lansing and the B. L. 5. prices for the U. S. city average (Tables 1 and Z). The least. squares regression of the panel series on the B. L. S. s eries {for Detroit), using the actual prices, produced a series of regression coefficients "b” as shown in Tables 5 and 6. In the tests of the regression coefficients of the food items in the ”cereal and bakery products“ group, the items of corn flakes, wheat flour, bread, soda crackers, and vanilla cookies were significantly different from 1 at the 1 percent level. The items of rice, rolled oats, biscuit mix, and corn meal were not significantly different from lat the 5 percent level. Among these items, only the prices of corn flakes in the panel changed wider than in the B. L. 5. series. The prices of other food items in this group changed less than the B. L. 5. series during the four-year period. The relationships between the panel and the B. L. S. for Detroit were thus similar to the relationships between the panel and the B. L. S. for U. S. city average (see Table 1, Chapter III). When dealing with the “meatsl' subgroup, the regression coefficients of the items were all significantly different from 1 except hamburger, veal cutlet, and lamb. Among the meat items, the prices IOf ribroast, ham, lamb,' and lunch meat changed more widely in the Panel series than the B. L. S. series. The regression coefficient of 1arnb was not significantly different from 1. The price-s of other meat 79 TABLE 5. -——Summary of regression and correlation results for cereal and bakery products using the B. L. S prices for Detroit and M. S. U Consumer Panel prices, 1955 to 1958 Level of Products I a? b tb value significance » Nope we Hmogumo v . v . NM- . Hm . om . wN . 0N . mane pmdom ofioo QN. mN. mN. «NN. mN. momsflpmpmoum Hmmdm CHOU mooo 004 v. w. 0N. Nam. oN. 5N. 5N. TVA moxwm CHOU N000. om . Hm . 0N . hN . oN . mgumong Nooo limw1mlu1l11|1ilm11m o .w -. WWNN ow ..N om. .N No .N mm1.N NM... MN. ”mg .1- 1111...... ....Na... - - 11-1%. ...3 1- 11m1m1l111Nm-.N1N11I11N1a-1a.a.-.11N.m1111a--mm.11NN1l.1m.m-...Na N.1.m.1.......wm ......s... axon. aw 3.0.30 oo .2: oo .oo_ oo .2: oo .2: oo .2: 00 dog oo .2: oo do“ oo .2: QOOh J44 $2 32 N2: 32 $2 33 32 3.15335 1.3% .ooQ .ooQ @3th powuom pownom @3qu owmuofiw mmou mo somudnfihamfifl 3:393 SN 52 £2 £2 52 $132 Eu: .8 .380 3.: 8N 4...: .3 .m.o.H.zvaofim Add sou .m .4 .3 p03 M33 3 some» 050: «mmpoe 1153 o» 228% “085980 .5 .m .Ev @284 um boow ~83 3 a. .32... 8...... ... 2...... .52 s as Ea 2 85:38:52... ......35. 2...... lillllllll . li“\\\\\‘\‘)1 out/How main—Quid: HdHfiquome< Ucm .U adom mva NOUNOU dMHHSmiUSH HNCOSMZ .mUquw—Uam HOQMJ. H0 SNQHDMH 4de 44’43 ’11.). sis..|rlu.”d“ \I’.‘ ‘L ,>|" I'll) 1 QHQQHOW . Mufimv mu m 38.5.5) $4.12 m- ,1 .N. mama... Agll'l 1 .ItnI‘I‘\1I - VIVAi- ‘1' Child .11- rI1IlIldC Ildlul11ll 1 snvslsssd f'>-.dl'-.IVVI fills-Ila... finF-fivfh11W-Iw RWAHAUH -d.wda.it nil lilwlln.‘|1l.l! 1|. Innis-shun c>dbsw-.IV~ 1|1lll'\ .1 x 9 pm .H a. .m N .... Np .a 3 .a 2. .N an .N 3 .a 5% .m 88m :S on .N m .N 5w .N i. .N mo .N am .N o... .N mp .N 3 .N 8am $12.3 SVNN .N .:. . 2. . E . A: . mm . 38.. ion 33 2. A a ..N m .N on .N 3 .N am .a S .N Na. .N am .N macro ion 88 .:. .m m .m. N .m 3 .s S .N. 2 .p Nv .e Na .5 S. .w mam. E. .N N .N .... .N 38.. .328 «N .a w . w . :62 n2 Na. .m c .m p .m E .m .33. venom as .w an 4. 8 .v 2 ... 2 a. a8... z... 2.33 No .v 8 .m mm .m .. 3 ..N N .m 38.. :4 $3me 2 .w a .N a. .N 2. .m an .m 2. .N m... ..N .aN .m we .N 33.5.9.3 83 em .: o .N N .N. mm .2 pN .2 on .2 ms .2 3 .3 5. .N Noam S. .3 N. .NN o .NN 8 .3 S .3 NN .NN 3. 4N mo .NN .:. .NN 332 a. .3 p .mN w .8 2. .NN S .:. am .3 ”N .8 3 .8 «m .eN fin a rayon :82 3‘ . 930 NN . NN . pN . MN . NN . 8:8 £00 83 E . S . 2 . I . E . 8% >226 23 em .N 3. . 2. . 2. . 3. . 3.. . panama. 63 umonm .munanmfiofl mmuvo S . a .N Ea .N 2 .N S .s 2 .N E .N so .N 339:. 63a _ 11.0280 23 ms . w . v . S . 3 . 2. . E . 2. . S J 28398 «pom 8.3 «m . Nm . wm . E . S . 3582 6.8m :8 $2 52 N2; 82 $2 $2 32 nonampnomxo 388 .ooQ .qu common v3.39 v3.39 vomnom ounno>m mmz mo somusnmhummfl 3533 3N 52 52 £2 £2 $-32 / 1 Eu: .8 .320 .m.o.H.zvoEofim ..mé .Se .91—.3 l1 l1 Um .D new .m .2 :3 080a um room :33 3 85.23:... 3.32; 2333383 982 “N 83 333 2......H 8.5.80 .s .m .2. “Edam ouguhoman grassed 050: um UOOH don—On 0» ooaduhomam 950.30% UOflGfi—GOUII .0 W438 nunvAu— can oavu an. .vva-dav rill-nv-n a... unnulu- na.. 01.. !\C I! . -..?! !I II c- - :4 v -........w -uxr..-.-.cv- mu ......d:-.:_.—.a .->_.-...-.uV— nv-.n..av!— dam unavavh 55.—53. Ava riUF-n...u-~0AhA-wtuv>m,.-,~...~;..~f~h. ..wlv.~...u.hah:n|~|-l-.n n~>-- ~..~§ u...:.:I',I-.¢Inur...,l [\I u-IuIIII\.I. 100 om . on . ¢m . do. . on . mafiam $3.3 mm . 3. 5. 3. 3. 2385 Toma... ma. 3. mm . 2.. 3. sum 8&0 $23 wm . mm . mm . on . mm . 33?. £th $33 up . w . m . om .m mm .m a» .m an .m mm .m SSH. «(:3 S. .N m .N v .N 2. .m 2. .m 3 .m S .m mm .m mm .~ "2.8. 8.. .w firm 3 . S . 3 . 3 . S . >825. ~-m-o~om E. . 53320 mamhroum S . - . om . cu. NN. 3383. Tmésm 5% .N ¢ .m 3... ... S 4 om .H 2. .H S .N S .N 238.3 a 20>...“ TMJSm 8 .m v .m w ... mm .N 9. .N t .N S .m on .m Anion 8.2. o. m. 3. pm. 3.. 2.... 3. McNamdmmm m. w . E . S. . mw . a. . a. . MwomMoIHMoMIonm vm. . w . N .H save “50:05; @0530 S. . 3 . 3 . 2 . 3 . «can. 58% .83m 33 2 .H S J 3 .H S A 2 A 33 38 .:.fio 2mm 3 .N H g 4 no 4 ...w . X. . i. . 8 .m mamofiom _ 3m mm .H o .~ w .m 3 .H 3. .H 3 ._ mm .H mm .H mm .. 33.325: E E E a 2833 3mm 8 .m . p .m o ... . . .w 8 .m a. .m S .m 3 .m S ... mamoylnwflm 3 4 333m 2 . S . 3 . S . 3 . 2.. 4 ammosmm $.88 32 $8 $2 $2 $2 3.2 $2 confignomxo 1.38 .ooQ .000 @3qu vomnom v3.39 vomnoa 0mmao>m mmz mo £0333»qu 352.3 .8. £2 £2 £2 £2 5.33 I! ll\\ 83. .8 n:55 m o H 2.2.2.. 3.: .:.. m A a / 1 Add new .9213 coguuomam 0>mfiw~0m 3&ch Hana—.330 .D .m .2» 080: um UOOH Han—o“ On moan—H09»: o>fld~om 38:33 3 35: 082 S 82 H283 Evkwflmfimm ooguuomfifi o>flmaom 325.80,: .o ”.39; Fvlvlvc nlfivlidnvdli wI-'.O 'iFlI-U-u tum -II\.F\.‘ .:.! o...n-.-¢ r1./av-..-1\d~n an v.-dad.nlv.n--u ruslai'u-rhV- —..-a- Aid hd.U-hnwfivh|AUR-FIIW mv>ifilfiinflml rvAhth-n u-y -uAvAvh uvnlunuuQvch-.v -\.¢ w-I‘IAI< I. m ... 4. . 2 . 2. 2 . 2 . 2 . .883 oNNm 1 v. 4.. N... . mm . S . om. . om . 3322230 22 ... 4V . wm . S . S . No . 5 . $2.33me 8% S .m o .... .w .m co 4. 2 4. 2 4. mm 4. Nm 4. MN .0 £32 .3239 mN .2 o .2 o .2 mm .2 3 .2 S .2 mo .2 2. .2 4w .2. $330 2. a 33.2 2. 4. 2...... 22:5 2. .m N 4 o .N S .N 8 .m S .N 2. .N 2. .N on .N .835 SS 2 4 4N 4 mN 4 mN 4 mN 4 oo .N «$24... 02380 82 2 .N TN 0 .N 2.. 2‘. cm. Nm. om. 4.82.85... 6mmmmoon90 omooAD own: .3. 3. 2.. 2.. 22.. 45.2.2.8... . . 233mg. 38.5 22 NN 4 N 4 N 4 mp 4 mo 4 2 .N X .N ..N .N ummmxuaéum 39.5.3 009 ommm Nm . Nm . mm . 2. . 2m . E38 8200 22 2 . 2 . 2 . 2 . 2 . .3822. 633303 .22 22 3 . m 4 N 4 2. . 2. . N4. . 2. . Nv . 2 .m 332323 .22 22 A202. .2 354. .m N .m 2 .N 2 .N on .N S .N 2 .N 4.3 34%.... £82 .22 8: :co .m N 6 3 .w 2 .N 2 4. 3 .m S .N a .2.» .43... $22 2: ...N .2 N .2 4. .2 E .2 3 .2 3. .CN mm .2 mo .oN ¢N .2 Euscoumifim mm . 4. . w . mm .N 3.58 582$ oN . m 4 8:N 4 2w 532 .4030 S 4 fin .:.?Q 22 $2 $2 22 $2 2.2 $2 8.32.89... 1.38 .000 .ooQ 63.39 63.39 63.39 603.89 owmu0>m mmo~ mo 605352.3me $422.84 new £m~ 4.3m“ Jumu in; Nmnmmw~ . 80: no 93050 (\I“ 08 / o .m.0H.E 9.3m ”.28“ ...mgmv l Awfiuom .3443 ...No98: 0N: Ifbdfivd ooguuo9gm 933303 a 0503 ad 600m .9430» 0» 305% ngucou .D .m .20 ogon ad 600% flan—Cu Ou oodduhomflum 9’32?“ 62.5280..- .o 39.2. 050: «m 600.“ 230» 3 0323.399: «33.36% U! -V I -II\ I Ida's-U flinch-01'1- r1.vn-O-.v ll-F-Avn- Jaw .Iacuunn-c rI.)¢Ulv-IIVI v! vfinlad.I!|n-nI-— II>1UIw-:I- uUAvau -w-Avd Av. F-IIlIIxAI [Ilil‘lll uI r!.v--.h,-A-a~.-uq nv.\.r-h-:.~.-VV\\ IilIIflU-v- .Oll .IFVIU.‘ -.14.IAU..I AU. cw.i-I£UI~AVAQII..:‘ m'Wf‘iINQOOVQ ‘ IIIIIIOIOIIIIO I . \: tlleI’R.‘ 102 . o2 . 22 . o2 . o2 . No . 3.2.2322 oNNN N» .2 N .2 o .N 2 . 2 . NN . 2 . 2N . NN .N 92.22 .33...qu NNNN N: 222 3 . 2 N . NN. . NN . NN . N: .522 £8389 2NNN NN. N . N. 3 . S . NN . Nm . Nm . 2 . .83an :2. Ne . 8 . No . No. No. ......N 6.820 2 . N . N . NN . NN . cN . NN . NN . 33.8 £825 :2. 2 . N2 . 2 . 2 . 2 . 33835 2N3 No . 8 . No . No . No . $3322.60 3:. ... N . S . No . S . No . 8 . 88822 “Sam ENN Ne . No . Ne . Ne . S . €33 NNNN N2 . N2 . N2 . N2 . 98.2.2.9 oNNN N2. 2. 2. N2. 2. Eco NNNN 2N. N . N. NN . 2 . N... . NN . 2N . 3880 ENN NN. N . N. No. 3. No. No . 3. SEN ..fimm NNNN 2.. N. N. 22.. NN. 82 2.. 2.. No.2 83.3 2N2. NN . N . N . NN . N2. . NN . NN . NN . 2.2.0 83 NN . ... ... 2 . cN . 2 . NN . cN . cm .2 232.0 2222. NN .N o .N N .N cm .2. NN .N NN .N NN .N 3. .N 2... .N. 22.22 6.228%; N2. . N . N . NN . NN . 2 . NN . N2. . 3.2.89 onm N . N . 2 . NN . NN . NN . NN . 82.8333 N2NN NN . NN . NN . 2N . NN . oN . 83.52.26 5.23350 :3 N . N . NN . NN . NN . NN . NN . 3.220 NNNN N2 .2 N. N. NN. 2. NN. NN. NN. N22 355m 3.2 No .2 N .2 N .2 2. . 8 .2 N2 .2 2N .2 NN .2 NN . 32.2.2... 28 N .2 2 .N N .2 NN . NN . 2N . 2N . on . 3 .N . . ..chao oNNN $2 32 N22 22 $2 22 $2 .2322..me 2.38 .ocQ .ooQ 63.39 63.39 63.209 63.39 @9395 mm: .«o moasaubumfl ages... .2 ...: .22: .2222 .25 3.3: /I.I..\\ a... .8 .286 .m.U.H.zV Ogogud A.” .D ”0% .m ...H .mv X I A.m .5 HO.“ .m .2 {v 600w130u0u 03.223 030: an 600m gnu—o» Ou- auondnm HQSQGOU .D .m .20 0504 am 600.“ Run; Du Ekrflmaom voguuonmgn 9,6330% 080: «d 600.« aduOu 0» 0382:0983 o>3m~0d 0053.39.23 0.963626% 609G3GOUI: .0 Waugh. .f-awVO-Iw/nV‘Vll-n'a‘ .191 l “I! I. ll ‘1 Ix! I . ...-... acqv.- ... -v .:.--u ruin-nv-a Ina .vaand. dd-dfiid Avd Inlalaa-a :v\iv¢lt-.v'l ul vJ-I~I.I-un-a-i rv>ndnuuriyh A—nv-nuwnn L.v---!I.-aun\ ‘— .fu ‘1‘ s o . n. - gun ’I\ u r; ot~ (ban-Ai-n d-w -.VA.Avu I-wunuu A1»,- nUMi-ndfldvdAUnh-in!ri>1flninmvlyf MH..i--.--5A a .‘ .~ N F 11.0IoIIlIII-I I lit ...II.IIII.\ fi 3 m 2. 2.2 . 2 . 2 . 2 . 822.2 3...th 2N2. NN . NN . 2 . o2. . NN . 3259 . 922950 0222209. Mnmhwv NN. N. N. 2. 2.2. 2. S. 2. 33.602. N222. NN . NN . NN. . 2 . NN . 3222.2 23882.0 N-NNNN o2 . 2 . o2 . No . No . .2822 2% N2. N2 . N2 . N2 . N2 . 5:252 .2822 N232. NN. NN. 2N. NN. NN. 32.23 68222 $232 .mqmvm mugs? No. No. mo. 3. mo. 2.2.2.2859 N522. S. N . N . NN . NN . NN . NN . 2N . em .2 989 2-22. 2. N. N. 2 . NN . NN . 2. .2 . 2 .2 980 N-oNNN No . No . N22 . No . No . 38.80 22.2. NN . NN . NN . 2 . NN . .2822 gnaw N-oNNN ...o . No . S . No . S . 2822 N824 N..2NN 2. .N N .N N .N 3 .N S .N NN .N 2 .N 2 .N S .N 3.22232; 2.2.5.0 NN.. N . N . _ - . a ... . . . 82.2.2 2:20 N . N . 2 . 2 . NN . 2 . 2 . 2.822208 222.22 2.222 N . N . 2 . NN . 2 . 2 . NN . m2.2.2.2329 N52... 2 . No . No . No . No . 25833 .389 22222 2 .. NN . 2N . 2N . 2 . 98.9 NNNNN NN . 2 N . NN. . NN . NN. . mm . NN . 3.2989 $2 2 . 2 . 2 . NN . 2 . NM2,220 T22 NN. NN . NN . NN . NN . 3.22.2229. N $222.22.. N222 2. .2 N .N N .N 2. .N NN .N NN .N N2. .N 2 .N 2:222 2.22280 22 22 N22 N22 22 22 22 3232232228 2.88 .039 .qu 63.209 63.39 63.209 63.209 032.295 mmm~ Mo "Saanzumwg 332819 .23 £m~ 5m— 5: fim— hmammmn IIII.II..\\.\\ 823.2 no 930.20 cm ooofion Geog “d 6009 Mayo» 0..— 0053 o .. .2099: fin .9 HO“ .m {H .mv 592 am 603 N33 3 L333“ 005320953 932.399 6092262900.... .0 NJQH Sognn Hoggou .D .m .2. 0802 Ho 600m 133 0» 00282309223 o>fld~3~ 080‘ a“ ”COM HNHOH OH ..m .5 now .m .2 JNV 2Inv-v- Ia. oauvnv- .I van... IVs-Invfl- .:.. vulullu- an... C!!! IV! II II r~.I-I..IHIIFUFI-.II‘ ..UUv-u-wdninvuhphr-N ni>uhuw~:v~h whWih huuwvxfih .u..nlnnuo Iv>:uu.o¢vfiu ri ...-......unnft-n :I\I¢-I.-.~V- mini-Av-n HI.- uvAVAvi unwiAvd Oi IL. luntarvoittlhn:l . 1|... .‘I.IOI’\ I 104 2 .2 3:80 8 .2 N . N .2 NN . 2 . NN . 2N . 2N . 823920 N222 22 NN . NN . NN . NN . NN . 2.32.22.82.85 .328 35.2.8.2 22 NN . N . N . 2 . No . 2.22 . No . No . 2 .2 .23.; :80 2N2. NF .2 2. .2 N .2 2.2 .N N2 .N 2.2 .N 2.2 .N 22 .N 2. .N 2532.33 .0329? 3335 06:. 2.2. .N N .N N .N 22. .N 2. .N NN .N NN .N N2. .N NN .N 320.3. .N .22. NN .oN N .NN N .NN NN .N2 2N .oN .NN .3 N2. .NN NN .3 3 .3 ”NEE 3 .258 $2220 Np . 22. . N2. . Nu . N» . 2:225 3309 N-2.NNN 22. N . N . 2 . o2 . 8 . o2 . o2 . 2822 22.20 212222 2. N. N. N2. N2. N2. N2. 2. 35.29 N..oNNN 2 . N . N . 2.2. . N2. . 3 . 2.2. . N2. . ..... “32.2822, N 32.2.22 20.2.20 2 . N . N . NN . NN . 2N . NN . oN . .89 N-oNNN 2.N. N. 2. 2. N2. 2. 2. Baum N-NNNN oN . N . N . 2 . NN . NN . NN . 2 . 3.2.23Jom3 :3lo NN . N . N . No .N 2N .2 N... .2 NN .2 NN .2 33.2 23.20 N-NNNN 2N . NN . NN . 2 . NN . NN . 82.822322 N-o2N NN . N . N . NN .N NN .2 NN .2 NN .2 NN .2 322922 88.2.2 mm .u >232 630m 2. . N . N . 2.82 .33 2.32.22 22 2.2.2 22 22 2.2.2 22 22 8.2322522... 2.32. .00Q .03Q 63.39 63.39 63.39 63.39 023325 mmo~ mo ficasnuhmun— 3.2.5.2 .82 .22 2222 422 2222 2.N-NNN2 . . 822 3 .2580 o o O Ill-K N o 2.72. 2222222 3.: 22 .m ..2 .e I! 1 0923 333239 02823932 0222232 «8522 am 603 320» o» 00282392.»: 332239 - / :ofldnm HOSIQOU .9 .w .2. 2.252280... .2. 2239...». 7m .9 How .m .2 .dNV 0804.. 3 600m 2.30» 02— L u -l- Hadlk:l .:....u a... I...u..n.d inr-nv: dud nvnuah 152-30 00 au'ilhl III uuuuu v I- -:--..-. h >>>>>> .....u.....deH~l.JA-:. u .erMUI-U: -lthv-h UI 300m n‘rdh-d OH UVYHI‘UNDMEU “xvud'uul 91..» .:MIHJ 12.-3‘ austwututz 105 No .2 2. .2 N .2 N2 .2 N2 .2 NN .2 2N .2 2 .2 2.222222 2.32.22.22.222 22 NN . N . N . NN . 2N . NN . 2N . 2 . 2.N . .NNN No.2. NNNN NN .N N .N N .N NN .N 22. .N 2 .N NN .N 2.N .N 2. .N 3230 NNNN - No .N 22 .N Na .N 222 .N 2 .N 822 2 2N 2 .N 2 .N o .N No .N NN .2. 2.22 .N N2. .2 N2. .2. 2.N .N 3222.22 62v . 323300 NN. N . N . 2.282 .223 2822 N .2 NN . NN . NN . NN . NN . .225. 0223.32, 28$ 2. .2 NN. NN. NN. NN. NN. 2:8 338.2. N-N2NN NN . NN . NN . 2.N . NN . 2226. 228202.20 2.2 2.N . 2 . NN . NN . oN . x28 .2.2 32.2 $2 22va .2 ON . 2 . 2N . 2 . cN . 25m 2382 .N 2223.22.22 NNNN NN .N N .N N .N NN .2 2.N .2 .NN .2 NN .2 2 .2 32.82 2.23.2.2 22222.89 v . N. . 92330 03809. 0 . m . @3239 3325 m .9 N .3 3033a 2w 32322362200 2 . 2.2 . 2. N2 . 2 . 22mm 22. N2 . N2 . 2.N . 22 . 2 . 83m. 232 NNNN NN . 2N . 2N . NN . NN . N223. 228920 2N2 2 . NN . NN . .NN . 2N . NN . 22.222.222.280 N . N . 2.3.82. 222.0 NN . . 22 . N2 . N2 . 22 . N2 . 222222220 N-NNNN 22 22 NN2 22 $2 22 22 823:5...me 2.38 .03Q .03Q 63.39 63.39 603.39 63.39 oum.3>m mmo~ mo 985252.3me 332319 .23 26m— Emuw no 920 .va N2: 22: .2222 3.3: \\ o unavoHon 05038“ A.” op How on 01H my UOOM Hugo“ O» 003» vEO£ um 600w HNHOH O» :03“ vaufiou .D .m 9:. .Egflmavd 0032720933 023.3349 030: um 600w 390..— Ou 30323093 023333 II A.” OD Ho.“ 0m 02 4v 0804 «m 603 2.302 0» 005320983 02323333 6ofldmaflouvlu .9 3:9“ I. -‘ I l“ I . GOA-Iv. llaanvd!d‘d Iii-.0..- lv-l-Aia- ....v -VA"A1I dn'uav- I‘uAInI: \ . a roll-- .:.-1. I. ¢-11.-ylnun-I-a mv>¢-I..-vUVI n'lnhhin- Aid r.i.v-n.m1h.hhva.‘-;.I ri>i u-u n-v¥\ n-d A—Iv-uflv-n -IcI-I-I-lf-flu.\ h. .ld ‘I\ . . rsyv-.uh‘~.~.--I ov>~s-.~.t~\ v... .IAVA" II>IIII.IIJI..o 1 \. .3,IIII. I .930m 3 .3220 unguaxfimnh 6 600002209 h W “Va JndOH Gmoqo . HE mQH V22. .2 N 2242 .2522 Noxofimm $h0002w . 00H . . 22 N 022222 . Jam:N 2.522.. 2.0232222 32 9. 2N2 EcumN .UOHONVMHQU fimvhw figfifi .UOUMHQ . $900de .ANQ 2.2m 220N029 6G0 2.30.2909 £03000 .mSder $120 mu0>uh N m .3220 30.3.0029 £0350 ~00> 622.0 28.01; .603 $365902: N06nw 0802:3990 02.3 2.2m uCO0H09 me .2205 0004 “w 2. N2 . 2 . 2 . N2 . NNHH222a 3.2 .8 N2222.2.2.9 oNNN . N . N . NN . NN . NN . NN . NN . 532% 2.3320 23 N . N . NN . 2.N . NN . NN . NN . “382822822 mm .N NN. N.m NN.N No.N :a 2N.N 2N.N NN.2 Emma 2:. N. N. N2. N2. N2. 22. 22. NN.2 2:3 oN2N, N . N . Nm . Nm . NN . NN . NN . NN . 3:22 3:89 NNNN S . N. N. NN . NN . NN . NN . NN . cm .2 Nc2332. 232$ 2NNN 2.2 . N2 . N2 . m2 . 2 . $35832 NNNN N2. N2. 2. N2. 2. 22° Nnfloou o2NN cm .2 . N .2 N .2 NN . NN . mp . NN . 2K . NN . NNENNSNN 2.23%; oN2N N2 .2 N. N. NN.2 N... .2 2N.2 NN.2 NN.2 8.2 22.22.8885 oN2N mm .N N .N N .... NN .N , mm .N N» .N mm .N 2N .N NN. .N . ”N220 .22 38.2 NN.2 S2 NN2 NN.2 $2 NN.2 32 8.23359... 23228 .009 .00Q 602.209 602.209 602.209 602.209 0wmu0>0 mms .«o cougauuummn o . N 35022.04 .20.“ .2222 222: .222: .22: 3059” Roan—02200 .D .m .33 080: am 600% 130» Cu ouguhonmcflm 06523.90“ 3.33 If \ ..m .D How .m .2 .4v 0504 pm 600.“ HNuOu on NoqNNHomEN 033209 ”.0 .Hon Geogad émcp HOH .m .1— .mv UOOH dun—OH OH 0056 0803 ad 600“ #090» Cu Eirmadaod 0050H09Em 25009019 6052235200.. I .0 HAQH 80:. MO mfiohmv 107 CHAPTER VI THE M. S. U. CONSUMER PANEL RETAIL FOOD PRICE INDEX The Usefulness of the M. S. U. Consumer Panel Retail Food Price Index The index measures changes in prices of the foods bought by urban families in Lansing, Michigan. Since "price change" is one of the most important factors affecting the cost of foods over a short period of time, the M. S. U. Consumer Panel retail food price index provides a measurement of changes in the cost of foods purchased by Lansing urban families, from one period to another. In addition, since the index is designed specifically for measuring price change over time, it provides time-to-time comparisons in analyzing food prices. The annual income, based on the three-year average (1955-57), for the M. S. U. Consumer Panel families was $5, 452, and the average family includes 3. 2 persons. Use of the index to measure price changes for families other than the Lansing urban families will be appropriate only to the degree of similarity in spending patterns to the Lansing urban families included in the index. Since Lansing is similar to many other cities in the United States, the index, derived for Lansing, can be used for measuring food costs in many urban areas. 108 Limitations of the Index Limitations of measurement. --The M. S. U. Consumer Panel retail food price index is not an exact measurement. It is subject to the many kinds of limitations that are always in statistical calculations. One kind of limitation is ”sampling errors. ” The degree of error introduced into the index through sampling depends primarily on the amount of variation in‘price change that exists within groups of items. To gain about the same degree of accuracy through- out the index, therefore, the number of price observations obtained for any item is conditioned by its price variability and its importance in the total index. For example, prices of fresh fruits and vegetables, which are important in the family food expenditure, change frequently and have different seasonal patterns in different items. So to measure the average change in price in all fruits and vegetables satisfactorily, a large number of them have been priced in the index. Another kind of "error" occurs in the index because housewives who give information cannot report exactly. Some probably report lower prices than they actually pay, some higher, so that these ”errors of reports” may tend to cancel out. However, the expenditure data are compared and ad- justed in order to keep the effect of these "errors" to a minimum. For a description of adjusting expenditures and prices, see Chapter II, Adjusting the Data. Limitations in using the index. --The index is specifically designed to measure the average change in prices of foods bought by 109 Lansing urban families. The index represents all Lansing urban families, but not necessarily any one family, or small groups of families. The index ”market basket" is held constant so that pTlC€ change alone will be reflected by the index. Changes in the level of consumption are not measured. Also, to show changes in the tonal "cost of foods, " the "market basket" (including quantities of foods purchased) would have to be reassessed at each period. Analyses of Changes in Prices of Food Groups The M. S. U. Consumer Panel retail food price index was calculated for all food at home for each food major group and subg:oup. The indexes obtained are presented in Table 10, at the end of this chapter. The changes in prices of each food group from period to period are measured by each group index in the table. A graphic presentation of the index of each group is shown below. Each graph is followed by a brief description of changes in prices over a period of time. All food at home. --The prices of ”all food at home" increased slightly from 1955 to 1956 (Figure 15 and Table 10), reflecting mainly the rising prices for.”dairy products, " “fruits and vegetables, " and "other foods at home. “ The index number for 1956 annual average of 100.0 (base, 1955-57 = 100) indicates that the annual average price in 1956 was the same as it was in the average. of the years 1955-57 (Figure 29 and Table 10). The percent change in 110 «nag “nag onmu nnma #2 N22 2N.,22 2N21_2_IN2.2__ 2N2 22fi.fi42 N22,2N2,222 m22 542.d22»2N2JN ox Ii mu ILna 80m u. 80% .34 ..ll. 82 .1 no.— .I o: .l n: I o3 dossm usISssoo .:.m.: '2. us v03 :1 no nsouua 330.... susus>s on» an nouna .3 .9: Agannnnns VIC—van prices from 1955 annual average (the index stood at 99.1} to 1956 annual average was 1 percent (100.0 -- 99.0 : 1.0 = .0101 or 1.01 percent}. 99.0 99.0 Lowe-r prices of "meats, poultry and fish“ and "cereal and bakery products” were offset by price advances for the other three. reaps described above. As a result of increased prices of "meats, p‘eul'ry and fish, " ”cereal and bakery products, " and "Other foods at home, "" the annual average prices of all food at home advanced l. 7 percent from 1956 to 1957, despite lower prices of "fruits and vegetables" and unu- changed prices of "dairy products. " The 1958 annual average index at 103. 4 was up about 1. 7 percent from the 1957 annual average, and 4. 4 percent higher than the 1955 annual average. Higher prices of ”meats, poultry and fish, " and "fruits and vegetables” more than offset substantial reductions in prices of ”dairy products” and "other food at home" groups. There were two seasonal peaks in the prices of all food a‘. home. One occurred in the early summer and the other occurred either in the late fall or early winter. The lowest prices fell in the first and tenth periods in each year from 1955 to 1958. Apparently, these seasonal patterns resulted from the prices of ”meats" and "fruits and vegetables" (Figures 17 and 19). Meats, poultry and fish. --The prices of the "meats, poultry and fish" group followed a downward trend from 1951 to 1955, leveled 112 off from the middle of the spring in 1956 to the early spring of 1957, and then fluctuated upward steadily through the end of 1958 (Figures 16, 29, and Table 10). The annual average index differential between the highest of 1951 (125. 4) and the lowest of 1956 (96. 9) was 2.8. 5 pomts (down 22. 7 percent), and between 1956 and 1958 (110. 5) was 13. 6 points (up 14. 0 percent). Seasonal variations in the prices of "meats, poultry and fish" group closely paralleled the variation of meat prices. The seasonal peak in prices occurred during the summer months in each year from 1951 to 1958, with the low coming during winter months (Figures 16, 17, and Table 10). The prices in the ”meats" subgroup apparently followed the same patterns as the prices in the "meats, poultry and fish" group. Annual average prices of meats fell 25. 5 percent (down 32. 8 points compared with the annual average index) from 1951 to 1956, and went up 20. 9 percent (up 20. 1 points) from 1956 to 1958. Higher prices for meats, from 1951 to 1956, were partially offset by lower prices of fish in the "meats, poultry and fish" group; and, from 1956 to 1958, higher prices for meats were partially offset by lower prices of both fish and poultry (Figures 16, 17, 18, 30, and Table 10). The annual average prices of poultry were highest in 1951 (the’annual average index at 144. 7) and lowest in 1957 (the annual average index at 90. 2). The annual average prices of poultry declined 37. 7 percent (54. 5 points) from 1951 to 1957, 30. 4 percent (44. 0 points) from 1951 to 1956, fluctuated 113 upward only . 2 percent from 1957 to 1958, and went downward 10. 2 percent (10. 3 points) from 1956 to 1958. The annual average prices of fish were highest in 1954 (the annual average index at 105. 8} and lowest in 1956 (the annual average index at 98. l). The annual average prices of fish climbed up 6. 2 percent (6. 2 points) from 1951 to 1954 while the meat prices dropped 18 percent in the same period. The fish prices dropped . 9 percent from 1957 to 1958 while the meat prices went up 10. 4 percent (10. 9 points). During the period from 1951 to 1954, the priCes of meats and poultry followed the same downward trend while the fish prices fluctuated upward. The most important price decreases of meats more than offset the increase in fish prices. The prices of meats and poultry dropped only moderately during the period from 1954 to 1956 while the fish prices decreased substantially. And then the meat prices moved up again from 1957 to 1958, while the poultry prices remained unchanged after having a substantial decrease from 1955 to 1956. The fish prices declined only . 9 percent from 1957 to 1958 after having a 2.. 7 percent increase from 1956 to 1957. Dairy products. --The prices of dairy products were more stable than the prices of most other food items (Figures 16, 29, and Table 10). The prices of the I'dairy products" group moved upward slightly from 1951 to 1952., turned downward moderately from 1952 to 1953, dropped sharply in 1954 and 1955, and then climbed up again from 1955 to 1956, and, afterhaving a winter seasonal increase, 114 fluctuated downward through the end of 1958. At 105 the 1952 annual average index was the highest during the eight-year period. The annual average prices of dairy products declined 8. 7 percent (9. 1 points) from 1952. to 1955 (the 1955 annual average index was 95. 9, the lowest of eight years), were up 6. 0 percent (5. 8 points) from 1955 to 1957, and followed a 3. 2 percent (3. 3 points) drop from 1957 to 1958. Higher prices for all kinds of dairy products in 1952,) compared with their prices in 1951, were responsible for a 1. 9 percent (2 points) increase from 1951 to 1952. A . 4 percent drop of the annual average index in the “dairy products" group from 1952 to. 1953 resulted from the price decrease of ice cream and butter. Lower prices for ice cream and butter in 1953 more than offset the slight increase in fresh milk prices. The annual average prices decreased 5. 1 percent from 1953 to 1954, and 3. 4 percent from 1954 to 1955. Lower prices for all kinds of dairy products were responsible for the decrease. The annual average index (101. 7) of 1957 was the same as in 1956. This indicates that the annual average prices in 1956 and 1957 were 6. 0 percent higher than in 1955 and 3. 2 percent higher than in 1958. Higher prices for all kinds of dairy products, except a slight decrease in prices of ice cream and processed cheese, helped the 6. 0 percent increase from 1955 to 1956. A 3. 2 percent decrease in annual average prices from 1957 to 1958 resulted from a decrease of prices of all kinds of dairy products except evaporated milk, which showed a slight increase. 115 . hum.— onoa nnaa 22.222.22222 222222222222_222422222222_2222+2222222 on 1 co ‘Ksl. 62.11029... 0.. (.a’. 3033.. {...-ass .:.-2...... o. o. o o o ....o...4or..6s.o. 9”... .1); v. .\.>(. 82 ..K. 45‘ 34.3.1.9! .\ 303...: his: 1 o: o‘o‘ 22.: van .h—uassm .33: .... on.— OS ens nnmu «mad anua 2222.232222—222224222222_2224.22222N2 22222322222 ON 1 co 1.. \P .0... o o. .o 2.... 1.1.659... ..o. 9......- 8d i. .l.!. . o. o o o o. ..0 o 630...... t...o...o!s...o...o... {:63}... o .o l.’ V.‘..’..’.‘. .s...s.. \ .(. (.-.). Y.-.‘. .uoazuh ”u“! I Odd 'c‘ ‘.\. ...-a’o. .(. 1.. ...: 23.. 652:8 .3822..\ 11...... ...». .36.». .. l 22 ‘ (0 ‘o/o o’0'o‘s-s-s. .( 03 2821...-nno2vxo32 ~08.— uslsssoo 5.»..- .nuoavoua .033 can ...-«u use .5315.— .susol us 22:93 hen-I. noon no 003%. :30... snag; on”. 3 00:38 .3 .u: I... I‘lil s‘i‘t c .:.-(3 lishls' C‘s-VIIV.I 116 «mad nnau Amma 1222:2222—2222:2222: N2222LEN2222: 2222N2222222 I4 doesm sols-coo .:.m.! n.«« was .oucdl no unsouunsu poem «0 souuua 2u¢uou 0m0u0>¢ 0:0 cu oomcqnu 04 on 00 on on om cog odd end on“ o¢~ own “SEENNSES .2 .N: 117 mnaa sand mnbw nmm~ 222.222222212224222:2_222222Nfi2q2_WN2N2N2222212 oN nuns: II on In co II as I: on co Os... on so .- . ..0) . 0‘ M o. 0.51 ..0. 01 1.1 1. 0 d1... CON 0 O . so. 0. 0 so 0 o. o c. - O s o O O o. I c K o. . \ .s.).. . ‘ ‘.’.‘ I. ‘. e. v o .\ ‘ 0‘0 0‘ £.d~ 0‘ o 0’ \ ‘ 0’s ‘ .oo~ :1 Odd I1 cud II and I1 oea II and «again coacwuaoo .sa .wwh 118 unad _4«Aqawmfiwq2222dN2222mnm~222qafl222424N2nm~2212_ q__q—_q_4444 J 0", 2’ .\\>/ 2 \ I \. 2‘\\ s / \22 J ( ’2 \II. 2 I1 2 2\ 22 > \x /L 2<2 22 2.2 )2 . L x . . x 2 (\lx/ \\ a : ; \ 22 . .2 1 ’2 x 2 2 2 2/ N . 2 2 2 2.. 22 2 2 < 2 22 2 .3398 \q 22 222 22 2 222 x .L m 2 2 2 s 2.2 2‘ ‘ 2 Cc. o 2 yr 2 .. u .. \ "4 0°.— . l f" 54k oz‘. .2 2 ...K 2 . 2 . .2 ......e. .: ...-.. 2 . ...... .. .n s .. 2 .. ...r-)...... 2. 2. 2 . . I 32 Mn 71( o .....o. Iak . ... \ .... \ ... 2 .. 2 M... .... .0330»; 2.3.» ......r. 22 I oS . ..t o . ( nu. 0.. \. 3:32— 22:2; I 222 d 32 «n3 nnS 222222222222fl422.2.22222222%....w..q2.14742 oh .....0. .... "ta. \fluf, .o-Oo . .... 2... , . J on .... .0 . .0 \Q. ... .. u. .... . ... .. x- -I. 0.. \fium \wx 2. ....\ 2 .. x n. .1 2.. \ .22 22 .. \ 2 . ...... \ ‘\\.. .. . .... \u/ ‘ o. o L JH‘M 2. .. ... \ . . I... Ms. ...» .o 2 2 .. .12l .. ... 2 .. ......u x a. 9.: 2 ... .0. \ .... ’ of Q... ‘ ......u o 1.. \ n. 2 .. ..0. on .c 2 N. 22 2 \ . 72. I o: . x .3322 for. \2 . 2 22 - \ r ro~n3onu> 5.0.: 1 14L . 223.— ...-I356 .:.».x 3.3332, 22.3.“ vs. 312$ ...-am we iguana. voou no .033 :30... ouch—2; 25 3 usual-8 < “81%;“: 2:22.222 .ON .9: ‘ \IPJIIII ,l lli-l ..." ’ | .. Id ”It 0“ ‘U' , \ I‘ui IIII‘Itdtu Il‘h 1:- ‘U 1 uvlu‘ II‘Il-|.\ Iv III! I I I I1! Illllun'u...i 123 ana2 snma anaa 2 2 2» 2 2 2 .2 2 2 2 .fl,2 _ In a 2, 2 2 2 2 s.2 2 2 22 2 2 a 2, 2 2 2. 2 2 2 2.4 on IIOQ :333.» 23 2.} I.oa ....o.... ,2-..’ LAMB/:1!» 1...}: Yu‘.) o... )3. L“W.eh+hv.) SI k\b .A wo‘nn" I<)Ik 0*“ coo... ( ALA-..0... \ 0:, o. o. oo’OAooOoolooo o O .0000. :2 8“ nfpl$(t\..!1$§§.\\ \IIIK /l\vi"1/ ‘N 22.x .332. 2.2.2228 \ I o: coaa-uuaop an. nuasuu vogue IJONA .I one nwnH - nnaa 222q242 22242—22 224m074q2: 2—2H42q452fi22 On .332— .3530 \\. .1 ca .:. .- .. oo2 00.00.000.00; » .2) I... a: .22 \ x I .o2n-uguo> econ-o ~ . < ' l cud ...—nauouop vs. 3.3.:— v0.2.5 2839.23.35 223m hula-a8 .22.»... €033.30} vac 3.2.23 v3.5 van .3333: 35.3 .3393 22283 no 22:83:- 233 no .03.:— 33: 23 n.— cow—.28 .2~ .u2~ 124 mna2 sno2 ona2 2 2 2 2 2 x2 2 2 \q 2 2 x2 2 2 2 2 2 x2 4‘2{1fi\2 2 2 2 fifl 2 2 q 2 2 \q 2 fi 2 2 H 2 10". 3 ‘ l ’ . . o .. o I. n . O. .’ O... s ..0 . . I. 10’. \ J L . o . a. u. at. o. O. I... )0 o If ..0 .:.....o ..0 ...- ... ... . 9‘ o «..0 0...”. 000 .0... a on... o . . O a. o. o. . on. o . .. 7 . .\ ... ... ... [k lbdld D. r ’ 0’43... brooquo cad. om ca 2 ou2nuh auuouh . .u2nduauo> nououh 002 022 cN2 on2 0&2 on2 nno2 «n02 22222222q22w_222fi2222222.2 om .~.. .... . \ ...-.0.. 0.. .... ......o.“... ..... . .0.. .0... ...... .... . ... ...fl . \ ..... F ......... .. 0.3.... . I .... .... . .1... .a...... .... H o ...... o... o O o . o’c .. .n‘ h pi! ‘ I . u 1E r o. 3. . ‘u. n? \ I . ’ \u . K . . W . .. . . > P." .0.... ....o.... l. . . 9.... :2 223%; :39: z s. . A . . . r o - o - o n. o . .U2nuh nououh |\\\q r. .\ . ‘ . .\ 062 an 022 282122.227... 2onca u¢ISuaoo .:.mex .002nnuou.> van .uuuauu nououu we unsouuaau voOw no noouun 22.9.u can :2 nouucau .Nu .u2u u¢o~2 125 frozen vegetables (up 1. 7 percent). The annual average prices of "fruits and vegetables" dropped 2 percent between 1956 and 1957, as a substantial reduction of all processed fruits and vegetables except canned fruits rose . 2 percent. The prices for fresh fruits and vege- tables between these two years showed little change on the average. The most important price decreases were for frozen fruits (down 10. 9 percent) and frozen vegetables (down 9. 1 percent). The annual average prices of the "fruits and vegetables" group advanced 5. 6 percent be- tween 1957 and 1958. The increase of 5. 6 percent in average prices of ”fruits and vegetables" was the largest annual advance of six years. The increases were due mainly to sharply higher prices of fresh vegetables (up 3. 6 percent) and frozen fruits (up 44. 1 percent), and dried fruits and vegetables (up 4. 9 percent). The'decreases in average prices of fresh fruits (down 3. 5.percent), canned fruits (down . 1 percent), canned vegetables (down . 8 percent), and frozen vegetables (down 2. 1 percent) less than offset substantial increases in prices of fresh vegetables and frozen fruits. Other foods at home. -—The annual average prices of the "other foods at home” group increased 1. 3 percent from 1955 to 1956. The advance of l. 3 percent in the prices of this major group resulted from higher prices in all its component subgroups except "cooking oils." (Figures 23, 26, 27, 28, 29, 3f), and Table 10). Higher prices in sugar and sweets (up 1.. 5 percent), partially prepared foods (up . 8 percent), beverages (up 2. 6 percent), fats and oils (up 3. 0 percent), 126 and miscellaneous foods (up 2. 6 percent) more than offset the sub- stantial reduction of cooking oil prices. The egg prices were unchanged between 1955 and 1956. The annual average prices of the ”other food at home" group showed little change between 1956 and 1957. The annual average prices of this major group rose . 7 percent from 1955 to 1956. However, the prices of all its subgroups except eggs regis- tered their annual advance. The higher prices for sugar and sweets (up 1. 5 percent), cooking oils (up 4. 1 percent), beverages (up 2. 7 percent), fats and oils (up 2. 2 percent) and miscellaneous foods (up 7.1 percent) were partially offset by a 5. 2 percent decrease of egg prices. Sharp declines in prices of eggs, fats and oils, and beverages, in 1958, were only partially offset by slightly higher prices for sugar and sweets, partially prepared foods, miscellaneous foods and the substantial increase of prices in cooking oils. The annual average prices of the “other foods at home" group in 1958 were 4. 1 percent lower than in 1957. Within the major group, the prices of 1958 declined 6. 5 percent in beverages, 4. 5 percent in fats and oils, and 5. 7 percent in eggs, while the prices rose 1. 5 percent in sugar and sweets, 5. 1 percent in cooking oils, . 3 percent in partially prepared foods, and . 1 percent in miscellaneous food, compared with the 1957 average prices. Cereal and bakery products. --The prices of cereal and bakery products were relatively stable and showed little seasonal variations (Figures 2.3, 29, and Table 10). Both cereal prices and 127 bakery products prices turned downward slightly from 1955 to 1956, and climbed up in 1957. The cereal prices were up slightly from 1957 to 1958, while the bakery products prices showed a.l percent decrease during the same period (Figures 24, 30, and Table 10). The changes of annual average prices in the "cereal and bakery pro- ducts” group were 1. 3 percent decrease from 1955 to 1956, 2. 4 percent up from 1956 to 1957, and rose only . 2 percent in 1958. The higher prices of cereal in 1958 (up 1. 1 percent) were partially offset by slightly lower prices of bakery products. 128 wnm2 ans 02.62 nna 2224222fi2142422222222122—2222fl22fi2222—422fi22222244 .unu a. coo. unnuo 32.2.3.2 hues-n 23. 2.0.2.0 ch mm on ma am no 82 5: O22 n22 6N2 ~oosu2n-nna2 a. a2 2oadu hols-coo .:.».22 .cloa u. .08 .2230 vs. .3039: P2322 vs. 2.83 me 22:95 no?! .08 no .0022. 2.23..“ among. 32“. 222 39.38 .2 .322 129 mnm2 sna2 ona2 nno2 _22222222:22_2142J222422242222422322122242224222224 on I4 on . . .d .... . ofioeo..o\pfoo.ooo 0.0.. 0.. ... .0... I? ......t’.. . ,H’. no.0... 0. .. O C . 0‘0'003'o. at 22 30:23....— huoxnn \ l 32 I. 022 I. n22 I. ON2 I; nN2 2832222225232 2.23.2 nuisance .:.».z .3260: 5.23222 23. 2.923 «a 3:83:- 632 no .3222 2.23: 3325 22... :2 39.18 .3 .3.— I “‘-I-\I'I.I 5' 'lIlI‘I I l’. I..- IIIVIU- .' 9 "Hi ‘ [III 1 .- 'u-‘l I‘U.-.I‘\I‘ 'I.‘ 130 .nu .uau , ono2 , nmm2 , ono2 nna2 _222222222222_2222222442222142222432222222222242222 on I. ma I. on L 3 .1 co !.0‘ .r.; ouoctm as. news» .) .>. I. no t . . o ). V.P.§...c..v....: v v...l.... V. .v. \. I. .).I. ..... 0.24.). . or} I o. . .... ...‘.o. .(co... ( ... a... o. a... 0.. .I(. .. or 0 b4 . .4 boo \ oil % (d P 4|. 4. . 4.) SM 2 .m . n ....n.a.:.. " .2 m a .\.\ a.... ’ ... ... . ......... .... .V ....).......V.). .. \v.4.\ ... .. ...... «.... 32 oeuauo>on .1 :22 n. n22 .1 ou2 J 22 20:12 unis-coo .:.».x ..uooa- van unas- van 333.»... «a 25822:. 38 «o noon:— 2238 3823 :2» 5 39.2.6 131 sna2 onm2 2212 ana2 22242222 2 II as anm2 , 22242222222222212211242232222222222422_2 I. an In nu II om 2. 3..» .p .. .. o v. . o . a . v. x . . r.) .... I. mm a. u I. .s I . .nms.o:A . h > A o‘..o‘ .‘. ‘ o. H ....O...O... .. 0.. ..0.... .o .. . r p. .. ... .... ....r...( .2 ...... .. ..< .. ... .... J. .....1.. \..~. ..q. ‘.‘...o. ..ul...04. (... .l o ...J... ... I ... ...... ... ... .. cosy ... ..d ....d1 ..N .. ... ... 8H .. ... .. ......... . . ....... ...... ......... ....... ... ..2 . 2 .. .... H H. . \ .. . ... l 32 . . . . .2233 2.0.2.98:— 5223uudm .... .. .. O. . . . r . .J l .. .2 .23 9.2208 \ .2 c 22 2 H ‘ C 'l 0N2 unannoua m22-2uucn van II nN2 2833-32 2.32.2.2 .oN .w2m 2.23m uolsnaoo .96.: .258 ..220 uc2xooo «o unsouunsu poem uo noo2un 22.00» ounuo>a «nu a2 uuuangu 132 cna2 rmo2 «n02 2n02 22fifi222222222—22112.222222_222222222222222222222222 .220 0:. cash \\\\\\ .. 0¢ 0m 00 0s 00 0a 002 022 0N2 0M2 002 0n2 2821.22.27.32 223.2 nuisance .06.: .220 v:- uunw mo naouunau .002 we noo2ua 22cuou oucuu>n «Au :2 coma-so .NN .u2h 133 «m2 sag onm2 nag 22222222222fi222+22222222fi2222222222222 222222222222 3 [on .14.... II. .P. ..I...‘). ..FI. . . .. ..b .I it (JO-.8412 8H IL 022 .220 6:. Iudfi ..l 0N2 I1 002 L 0.22 ..l 0n2 282nnunnm2 0232.62 eo=a2ucoo .2~ .u2m 134 «n02 nn02 2mo2 , 1 - .2... _22222:22222222:2222222222222::22222222222222222 J 2 I 8 2.. . ......,...... .. J. a r . M. 1:: . n. W .35. 002 . . .. . . . ..... J o: .......... .. .2... ........ J 32 .wtka. w. h m u 0n2 0¢2 2ou.& hula-coo .=.m.: 0: 0n IL0n2 own. no aaouuaao 6002 we coo2un 22inch owcuo>n «:2 a2 nous-nu 2002‘s manna 2.3232 .QN .m2h 135 on m.— «mmofl _ .N. _ _ _ _ . o¢ on mama . “nag __ _‘__ q __ _ __,d d _ __ _ .x. ..w. __ d _ _. _ _\.A__ _ __ q ._E_ _ .4 _ IL I on 1 oh I on ’9. ...... . I0... 0. .0 o co. ./ ..\... . .... .... . ..... I. Go .r. it... . ... ..0. o... .0.. ..0 .I ...» . ....m 2: ......o. ... u... 1 Odd .uuuxxxx. .:. .4 cud .l 03 Gem 1 and Aoognsn.nnoavxona~ vaaaduaoo .o~ .mdu 136 mm mm om an on hm on an um um on an — . fl . . a d 4 ca . ..... . .... . ....... . 2: . .. ............. ..:::: l o: 1 as 3268.. I 2: «lo: u. vooh :< «Io: u. vooh nozuo box-n v:- dauuo Aua_MquwwwHuxovuu mm mm cm mm Qn Mn an no on nn #n nn Nn an an mm on ,nn fin mm «m an . - — _ a q . 4 4 A J 4 1 fi 4 . 4 4 on I. oo ...:.-..:.... ..:...: S... «I... ..1 .....1. ... .:.. . .:.. . x. .. . . .... cod .. . us .1 odd nuumxx. ou~nluo¢o> van ouuahm nuonvoum huqca can .huuasom .ouuo: .../IJ and and Acouflfinunnodvxovcu doc-m unis-coo .:.mi coach» v08 mo .03.: :33 «mun—2rd ...:..:. «Au .3 nous-no . o~ . nah 137 um um cm nn en an an mm on mm mm mm mm mm d a a _ _ _ _ _ _ \/ ./ .\ . 1 o: 1 OS .0333; v... 3.3.: v3.5.3. voou cowl—cum hZ.«qu~.:..:. coacugonazi; hung—an ........ .l 09 .33.»; :39; r... 3: mud—coo II 3026 vs. unnamll .:.qu |.9| 3.3.; nououml 03 35 on mm on an «n ..n on “n on an .3 m. an an on mm m nn an an on «n on mm «m mm «m .n _ _ _ — _ a _ nu... . J1... ...... _ _ . . _ ca .... ...L + ... . v. 1 J11. . I u .a...‘ . \ r. v1lwcxlj I: L.......h.......§........u.o..1 ...d.... ...\. .. . 113......11/5. .ll. . A.... Ills .lv......... 8.— 0.. ii” ...“V. \ ....... ...% \ u n O. u . l k .... ht \ .... / o- .33.: v0.3.0 ..:.. nuansomo> . .... x. 1 ON.— £.uu~ ..:.... 1 ... /o .... ..:.-.:: a c.3332, 1. o .9 .... 02 6.530 9...! .30 van much ..I .:.. '35.; .39: u . 9 .I an: .. 3036: 3:5 .933 ...- o3 auuaaom !:::. x Aoofikn$n2vx35 ...:..— uolsuuoo .36.: agendas. team we .03.:— 33: 332; «use: .5 3 noun-no .on .9: I! F\ 'I f! .. I II I .lll\\|’\.l . . «.1. vii ...- II VI. - ...Jsd.l .1 nun-19v nrlld v:I-.v.ll .:.: ant! ri-VII A .v.u.'~.n-.. .NOH 0.2: MJVOH ©.NOH «INCH Ndom o.No~ .~.mo~ wéou 0.2: 0.2: ofio 0.00 ~.mo~ mvoow mom Ucm gm?“ Nléumfi m .WNH m. .mNH w.omH N.MMH o.o- o.mm~ mJVNm ©.mm~ ~.mm~ N..om~ $.N.m~ o.om~ ®.m¢~ >HSSOQ m.mod m.mm Nam m.m> N.©: mdmfi m.wNH m.N: 04‘: N.mw ed: odon m.mN~ N6: mm; .nEmA deN 5.6: w.m: m.mNH NJVNH v.0: deH NN.: 0.90% NANN ~.N: ~52 m.mm~ 56: 18> ad: 0.0: arm: ox“: 0&2 N.m: OJ: aim: NIH: 05: N..m: $.22 NJ: 0.4”: $de .850 m.mH~ N.OHH m.m- Ndmfi OJNM wfimfi ~.©- ®.N.Hn m.m- o.N- NJuofi o.mo~ N.N.o~ m.N.o~ xhonm “find odmfi ¢.mNH cde odmN 0.3: odm~ 0.0.: 0.3; w.om~ wdm~ o.N.mH N..wm~ 0.3; . .«mmm 55.2 «.m: mg”: N52 Tmfi 9N2 m. .a2 .62 5mm“ Q62 NAM: v4.2 .52 «$2 33: NANH N6: H.N.: $.0NH NdNH odma mimmm oémm «admm o.mN~ NdNH w.om~ m.om~ m.m- :md flaw Nufidom .mummét‘ unuumm®_unu- mlofifi ¢.N.N~ N..N.MH m.mm~ w.NMH v.oN~ wéma MAMA w.mo~ N.oo~ ¢.oo~ mwwm >6: 9;: m.oo~ m4: ed: N..on «yd: m5: mama N..N.- odmm mic vcm 3mm 0.2: m.mo~ m .mom >.NoH o.mo~ 5&0“ 0.2: w.~o~ ~.mo~ N.mo_ 0.9: muudwoum Knfimm e do 0 .oo m .2. m .ww N. .NoH N .9: a» .2. v .2: N. .N.o HEN. o .2: mUOOw mom vcm 33m N33; N. .2: 0.0m” Ndfll m .omH ¢.mm~ m.mmH o.Nm_ v.3; m .mm— wému >539”. m .om~ m .NMH oélm w.o- ©.NH~ $60 ©J- m .om~ v.9»; N.®- m.ON~ mm: 5,984 méwa "dag N.@N~ mdwa w.wNm oéum mJNH N.N.N~ o.m- odmn néau fimv> ~.N: 04m: H6: N..m: N..m: m4: NA: w.m: Nd: ed: xxx: madman umfiO m.w: N..N: MAN: 56: N39: mdNH NN.: 0&2 o.w: N.¢: $5: Muenm w .N¢H v .mvm @ .mwm N .mwé N .mvm «a .Nva m .Nwl N. .Mw; m .an o .N¢~ N .0“; momma w.w- ofimm vdmn odmm v.0mH a .ONH odNH NJMH N..wN~ $.N.N~ ©.©N~ man—mm: vfifi mama m.m- . m .mma mdma o.mNH vdmu muoma m .mmu ¢.mNH .o.vN~ gmm wad . Nah—30m £322. . unnuammaung- wwwhvxrd mm NH «A Ca 0 m \- O m .I . v m 3:on . 328.... 85¢ :33 v N ~ we ..N ”U00“ fivvuu o u I III II 30w no .9365 adouwQSm cam msowmaanxhwanmwwwua . 600 H N H So» $ch 355980 .3 .m . . 2:9 Cd fiJQCq 139 0032: x003 v 825.80-.. .2 mqm.y:fim 0.:0: N.N0: 0.N:: 0.00: 0.00 0.00: N.00 h0.:0: 0.:0: 0.00: “0.00 0.00 0.00 0 .00 000: mom 0cm. gmflr. 0.N0: uv.0N: 0.0:: :.0N: 0.N0: 0.0N: 0.0:: 0.00: 0.:0: 0.00: 0.0%: 0.0:: 0.00: 0.00: 03:30n 0.N0:. :.00 0.:0: 0.00 0.00 0.:0: :.NN: 0.0:: 0.00: N.:.0 0.00 N.0:: 0.:V0: 0.00 00:..QEm1: 0.00: 0.00 «.00 N.:0 0.00 :60 0.0:: 0.00: 0.00: : .00 0.00: 0.00 0.::: 0.NN: :mm> v .00: 0 .00: 0 45:, 0 .No: 0 .mo: 0 .00: :.00: N .00: v. .:Vo: 0 .:Vo: 0 .00: 0 .00: N .00: v .00: game .550 0.:VN: 0.:VN: 0.:N: :.0N: :.:u0: 0.N0: 0.00: N.NN: 0.00: 0.:VN: V.0:: 0.0N: 0.0:: uv.00: xpon: 0.00: :.N0: 0.00: 0.00: 0.00: 0.00: 0.0:: 0.00: 0.N0: : .00: N.00: :.00: N.:u:: :.0:: $000: 0.N:: 0.00: 0.00: 0.0:: N.0:: 0.0:: 0.0:: 0.0:: 0.N:: 0.00: 0.00: 0.::: :.0:: 0.0:: ”my? 0.0:: N.0:: 0.00: 0.0:: 0.0:: 0.0:: 0.0:: 0.0:: 0.0:: 0.N:: 0.N:: 0.N:: 0.0:: 0.0:: 0m: 0cm #55300 .333 ----mmo_---. m.mo: o.mH~ ¢.0NH o.m~H o.om~ N.NNH o.w- w.~o m.ow o.©w “.ow o.vw w.wm m.~o~ mwmm 0.:0: 0.N0: 0.:0: 0.:0: 0.N0: :.N0: 0.:0: 0.00: 0.00: 0.00 0.00 0.:0: 0.N0: 0.00: 2:0 :35 mwmr. 0.00: 0 .00: 0.00: 0 .00: 0.00: N60: 0.00: 0.00: 0.00: 0.00: :.:~0: 0.:u0: 0.00: 0 .00: RESUOPQ Nu..me vmmkmxfiw 0: N: :: 0: 0 0 N. 0 0 v 0 N : mQ§OH0 UOOr. 130$de _ li‘fim . ..0 Fr.\.ll.l.. ... mumfiICAVx 140 I.“ III 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.00 00mm0000> USA». mewfiab 0.00 0.00 0.00 0.000 0 000 0.000 0.000 0.00 0.00 0.00 0.00 0.000 0.000 0.000 mmwm. 0.000 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.00 0.000 0.000 0.000 0wwv000 mzw0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.M00 00000000g00umm 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0 000 0.000 0000 000 000 0.00 0.000 0.00 0.00 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0000000 0.000 0.00 0.00 0.000 0.000 0.00 0.000 0.000 0.000 0.00 0.000 0.000 0.000 0.000 000.0:020 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.000 0.000 0.00 0.00 000> 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 00x20n00000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0000 0.000 0.00 0.000 0.000 0.00 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.000 000m 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 000030 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0000000 0.00.0900: Jammy. ----0000---- 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0010:0000 .0 00030 000000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 001030000 00000lo 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 90000000000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 000030000 @000qu 0.00 0.000 0.000 0.00 0.00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.00 002500000Q0u 000.002... 00 N0 00 00 0 0 .0. 0 m 0. m m 0 000000 08.0 00005002 00:02: 20va 0. llllllllllllllllllll .ll 00300030000 .. n .0: Mdmaflfi < 15h.v>d 0 A M‘- M n Ayn N0 x N. AV m w.‘ M M! N EANZAVLH‘ xunv. al.0000000,... «ufiewuuuva~ 1:3»)? w‘ .u.In-.-‘.--vn5l vii. HVQ.IJI<:I. 141 0 .000 0. .00 0.000 0 .000 0 .000 0 .000 0 .000 0 .000 0.000 0 .000 0.000 0.000 0 .000 0 .000 05000.0 0.000 0.0.0 0.000 0 .000 0. .000 0.000 0.000 0.0.00 0.000 0.000 0.00 0.00 0.000 0.00 000 0.800 0 .00 0 .00 0 .00 0 .00 0 .00 0. .00 0 .00 0 .00 0 .00 0 .000 0. .00 0 .00 0 .00 0 .00 0005 0.000 0 .00 0 .000 0 .00 0 .000 0 .00 0 .000 0 .000 0 .000 0.000 0 .000 0 .000 01000 0 .000 00008 .8005 0 .00 0 .00 0 .00 0 .00 0 .000 0 .000 0 .00 0. .000 0 .000 0 .00 0. .00 0 .000 0 .00 0 .00 000000 0 .00 0. .000 0. .00 0 .000 0 .00 0. .000 0. .00 0 .00 0 .00 N .00 0 .00 0 .00 N .00 0 .00 00000 0 .00 0 .00 0 .00 0 .00 0 .000 0 .00 0. .00 0 .000 0 .000 0 .00 0 .00 0 .00 0 .00 0 .00 00002 0. .000 0. .00 0 .00 0 .00 0 .000 0 .000 0.000 0 .000 0 .000 0 .000 0 .00 0 .000 0 .00 0 .00 00000 000l 0.00.1000: .3002 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.00 00000000 . 0.00va 0.00 0.00 0.000 0.00 0.00 0.000 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.00 000000 0 .000 0 .00 0 .000 0 .00 0. .000 0. .000 0 .000 0 .000 0 .000 0. .000 0 .000 0 .000 0 .000 0 .00 000000000001. , -0000 000.80 ----0000---- 0 .00 0. .000 0 .00 0 .00 0 .00 0. .00 0 .000 0 .000 0 .00 0 .00 0 .000 0 .00 0 .00 0 .000 0000000000., .0 0000.00 0005 0 .00 0 .00 0 .00 0 0.0 0 .00 0. .00 0 .00 0 .00 0 .00 0 .00 0 .00 0. .00 0 4.0 0 .00 00000000000 monoun: 0.00 0 .00 0 .00 0 .000 0. .000 N .000 0. .000 0 .000 0. .000 0 .00 0 .00 0 .00 0 .00 0 .000 0000.00 00000.0 0 .000 0 .000 0 .000 0 .000 0 .00 0. .00 0 .000 0 .000 0 .000 0 .000 0 .000 0 .000 0 .000 0. .000 00002000., @20ch 0 .000 0 .000 0 .000 0 .00 0 .00 0 .00 0 .000 0 .000 0 .000 0 .000 0 .000 0 .00 0 .00 0 .000 0000.00 000000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.000 00000 60303003, 0 .00 0 .00 0.00 0 .00 0 .00 0 .00 0. .000 0. .000 0 .000 0 .000 0 .00 0.0.0 0 .00 0 .00 00030 0000.00 000.32. 00 N0 00 00 0 0 0. 0 m 0. m m 0 00006.00 0000.0 035:4 @0002: 000m? :0 i‘II‘III ‘ 1““! ‘II.\“|||\‘I"J |||‘l\’u‘|‘ . 1" 31:1. :30. ... a .N 0 0 0 C 0 O 2 xx 0» m v m. V ~ $4030.01 0 nun-unpnl" \ll 142 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00 0.00 0.00 0.00 0.00 020000005000; 0.00 0.00 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 msowqwzoumfls 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.000 0.00 0.00 000m 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 mzo.00300 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 000000>0m 0.00 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.00 0.000 0.00 0.000 0000000000 umpa 0:103:00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 00000::0000 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00 0000300000000 0.00 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 _M:cO0 pm @000 .SLWm 0.000 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 001::000> .0 000000.00000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 001:3000> . Cowoub 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 002:; 000000 0.000 0.000 0.000 0.000 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.000 0.00 001::000> coccmn 0.00 0.000 0.000 0.00 0.00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.000 9:00000cc00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.000 0.000 0.000 0.00 0.00 00000 .mmzmumwm; 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 00000.00cxr0 0.00 0.00 0.00 0.00 0.00 0.00 0000 0.000 0.000 0.000 0.00 0.00 0.00 0.00 00¢:3000> wad 003m 0 .00 0. .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 .00 .0 .00 0 .00 0 .00 00805.00 0.30m ~.000 0.000 0.000 0.00 0.oo0 0.0o0 m.mo0 0.000 0.00 v.000 0.00 0.000 0.000 0.000 00000 dmm wad swam I! N. o m an m N H deOHw UOOh NH 2 o— o w ommuokm ma mmmuo>m x003 v i 00:::<. . . . \‘ll‘tll‘ 109600080} .0: ”0.010028 I‘ll I‘I\‘I\I|i|ll\ 9.1 «a .H Aux/aw ad..1daur.< ”A I uynvflvd..-A~ 'cu..<\-\m~.. ”v. \0 143 0.3: m.wo~ 0.000 mdofi odofi 0.000 0600 0.000 w.mo~ Ada: N.mo~ v.2: .voH w.mo~ 00000.00 c0006.?“ N.NOH odoa ~.No~ wdQH odo odofi m.NoH ¢.¢o~ 0&0“ N40: H.mo~ w.mo~ N.NOH méofi @2055 _ .mvanmuwwmkz mJoH 00.2: 0.000 v.2: odd mdofi odo m.No~ N60: odo 0.2: m.oo~ méo~ «.do «50000000 .mfiduh 0.NOH @4000 0.2: Ndw w.mo odw 0.0: odwfi N03“ odoH m.mo~ m.mo_ Ndo mdo £00: .mwfiflmfiWMOK) 0.2: mdo vdofl 0.4m «tow m.mw 0.No~ 0.NNM m.©N~ ~.m: mdom odd mdoH oéo . 0.0mm: .mfidub 0:060 N.No~ m.mo~ oéo m.oo Odo v.2: 04.: 0d: >60: 04%: 0:2: MACH ado 00030000.?qu 00:00... midgb >40: mgvoa odoa @4000 v.3: v.2: m .moH NAOH o.oo~ odoH Ndo m do m .wo N.wo 000060.000 otmm ado o .2: N do w do 5 .mo o do m .mo m .mo ~.wo m .oo 0? do H do w .wo o .2: 00000.0 New 00:00 cwfih p.00: N.mw ado 0.No ado oéo 0.3: odofi m.m- o.m- mdw w.wo NA: «Km: >000donw m .Mo m do o .vw m .000 o .2: N .com 0 do m .oo~ m .mo 0 .oo 0 do N ..oo m .«.w m .ow mm: 50:00.4 o.mo Odo. Odo «two Hdo m .mo N do odw wdo w.mo odm v.2: odm o.mo~ 0003/ Ado ado “do «udo m .No .v.mo oxmo m .vo o.mo wéo ~.mo wdo Ndo o.mo 00.000090 H.050 N.mo mdo odd 0.000 N.No~ 0.2: o.mo Ngvo wéo~ wdo N..oo wéw «too mdw xhom m do «ado 0.Nom o.wo vdo N. dog o.wo vdo odo w.mo 0.No wéo add m .No Momm— odo mdo odo 0 do H do m4“: vdo odo ado ¢.mo ".No odw Aw.~d méo 0000002 oxwo M .No w.wo vdo odo vdo odo Odo w.wo Néo odd N.No Odo oéo 000$ @501 opfidom .3062 0.00 0.000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.000 00000000. >memm 0 do 0. .000 0 do N. do m do 0 do o .wo m .wo 0.02 m do o .wo w .000 m .mo 0 do 00.0.80 0. do 0 .000 0 do 0. .oo .0. .oo m do 0 do 00 do m .mo 0 do m do o .000 o do 0 do 000060.000 >000 L uxmnruwfimmpoo ----0000---- 0.00.83. 00 00 00 00 0 0 0 0 m 0. m N 0 00.08% 00000 13:20.40. UOwHMwQ #003 V 0000050080-- .00 waged. .:..IN 4 144 N.mo~ 0.000 0.000 .o.ooH 0.000 0.NHH 0.0”“ 0.00” 0.00H ~.moH 0.00 0.00 0.00 0.00 memmsL 0.NoH 0.000 0.000 0.000 0.000 0.000 0.00_ 0.00_ 0.NoH 0.00" H.00 0.00 0.00 0.00 00a 0:0 omfidom .6252 0.000 ~.moH 0.Nofl 0.00H 0.NQH 0.00H 0.000 0.00” 0.00“ 0.NOH ~.No~ 0.00 0.00 N.oo~ 00000000 . khwxmm N.HoH 0.000 0.00H N._oH 0.00 m.mo~ N.mo~ ~.~o~ 0.N0H 0.000 0.00" 0.00 o.oo~ v.00 ammhwm 0.000 0.NoH 0.NoH 0.000 0.000 0.00“ v.mo_ 0.N00 o.~o_ m.mo_ 0.Ho~ 0.00 0.00 0.00 muoswopa.>po $33 ”0 4.00.0an ----0000---- 0.000 H.000 0.00H 0.N0H .N.00 0.00 0.000 ~.No~ o.Ho~ 0.00 0.00 0.00 N.w0 0.00 mayoge00000030 0.00 0.00 ~.HoH ~.m0 0.00 0.00H 0.00 0.N0 m.~o~ 0.00 0.00 m.oo_ _.00 0.00 msomqmzwumsz 0.000 0.00 0.000 0.000 “.00” 0.NoH ~.moH 0.00 0.00 N.00 N.oo_ m.~o~ 0.~o_ m.-_ 000m 0.000 n.0oH m.~og 0.000 0.mo_ 0.00 0.00" 0.00“ m.mo~ ~.~o~ 0.00 ".00 ~.00 0.00 mzo 0 mgrm 0.00” 0.000 0.000 0.N00 0.00H 0.N0H 0.HoH N.H0H 0.00 _.00 0.00 0.00 0.00 0.00 mommpm>0m 0.000 0.000 0.00 0.00 0.000 0.000 0.000 0.00 0.00H N.mo~ 0.00 0.00 0.00 0.00 00000 magma Iona .333th 0.00 0.00 0.N0 ~.H0 0.00 0.00 0.00 H.00 0.00 0.00 0.00 0.00 0.00 0.00 mwuwwchoou 0.000 0.Nofi 0.000 0.00 0.00 0.00 0.00“ 0.00 0.00 0.00 0.00 N.00 0.000 o.ao~ muomamuwgmmsm 0.000 m.~o~ 0.NoH 0.000 0.No_ ¢.~o_ 0.00“ 0.00 ~.00 0.00 0.00 0.00 0.00 0.000 .:co; 80: pm 008 .5sz 0.000 0.00 0.000 H.vo_ 0.00_ 0.000 o.0o~ 0.No~ 0.00 0.00 0.00 _.m0 0.00 0.00 00::3000> _ .0 33.: Bio 0.NOH 0.00 0 Ho” 0.00 ~.00 0.No_ N.mo~ 0.00M ~.mo~ 0.00“ 0.00“ v.oo_ 0.000 w.~o_ mxznmuomo> CQNOHrm owmhm>m _ l. 3284 2 .2 Z 3 o m .0 o m v m m ~ 350% week vowhmm M003 v 1|! UwDQSCOUII .O~ Mmdquwh. . 3.1 5.- .v>4q .:.aua--< A»— I —yAuu-puam ,0oaa-1-unn». I ¥—.V0ds.(l ‘1 ..v.’\ 145 0.000 0.000 0.N00 0.000 0.000 0.Nofi m.~o_ 0 _o_ 0.00_ 0.00 0.mo_ Mofiofi 0.000 0.00 muwmemuwgaugm 0.000 0.NOH 0.000 0.000 0.000 0.NoH 0 000 0.00 0.00 0.00 0 ~00 N OOH o OOH 0.000 MFCOn 00 men; 00.:pH 0.00 0.00 0.00 0.00 N.00 0.00 0.00 0.000 0.00 N.00 0.00 0.00 0.00 0.00 00::xmmmp ”a 0000.0 @0000 N.¢0 0.N0 0.00 0.00 0.00 0.00 0.00 0.00_ ~.00 N.m0 0.N0 0.00 0.00 0.00 00::umwu> CwNOLm N.m0 0.00 0.00 0.00 0.00 N.00 ~._0 0.00 0.00 _.00 o.~o~ 0.NOH 0.00 0.Nofi muurawcwwokm 0.00 0.000 0.00 0.00 0.00 0.00 0.00_ 0.Nofi o.~o_ 0.00 0.N00 0.000 0.000 0.Nofi 001fi3000> . UQZCdH 0.000 «.000 0.NQH 0.00 0.000 0.00 0.m0~ m.mo_ N.0o~ 0.00_ 0.000 0.NOH o.~o~ N.~o~ msfishfluoccdn 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.000 0.0”“ 0.000 o.0o~ N.00H N.vo~ ~.00H £0000 .moanmumwmxl o.oo_ N.00 0.00 0.00 0.00 0.00 0.000 0.o__ 0.000 0.0H0 0.00“ m.mo~ 0.000 0.00 00000.0025r0 0.000 ~.ooH 0.00 0.00 0.00 0.00 ~.oo~ 0.000 0.mo_ 0.000 0.oo~ ~.~o_ o.~o~ 0.00 001::000> L 0cm mfidpm 0.000 0.000 0.000 0.000 0.000 ~.HoH 0.000 0.000 0.oo_ 0.000 o.~o~ m.~o~ 0.mo~ 0.00H mpuswo~0q00mmm 0.000 0.No_ 0.000 H.000 0.00 0.00 0.000 0.000 _.mo~ ~.oo~ 0.00_ 0.00 m.oo~ 0.NOH @000 . .mwm fluid Smflb ~.o0 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.~0 “.00 0.00 0.00 0.00 0.00 0H:aom 0.000 0.000 0.000 0.00 0.NAH m.¢- 0.NNH 0.000 0.N0 ~.mo~ N.00 ¢.~0 0.00 0.00 000 ataxq 0.0“" 0.000 0.00_ .0.oH_ o.-~ 0.000 0.NHH 0.000 0.m_~ 0.0“” 0.0HH 0.~o~ 0.N_~ 0.000 Hmm> 0.mo~ 0.000 0.000 0.00” 0.000 0.000 0.00” 0.000 o.mo_ 0.00 0.00 «.00 0.00 ..00 msmwcuumgxu H.000 0.000 0.00“ «.000 0.000 m.m~_ 0.N- ".00" 0.00" 0.mo_ 0.00 0._o~ 0.00 o.oo~ xyom N.~H~ 0.000 0.000 0.000 0.000 0.000 0.0““ o.0o~ N.moH m.mo_ 0.00 0.00 0.00 0.00 000m ommuoim 2 NH Z 3 o w 0 0 m w m N ~ mascam 000m 1333 @3qu Meg? v Hum35280.. .3 mqm/wm. ..A0mr-nvnd ......” f I \x v..~.u>> uAvI a... WAK--L ...! \VAY 146 w.oo m.HoH m.HoH m.m0~ 0.oo wu~0~ 0.0o ¢.HOH 0.m0~ v.oo m.oo 0.mo m.mo H.~0H woow .mmvm UCm fimflrh 0.00 0.00 0.00 0.00 0.00 0.00 0.00 H.00 N.o0 0.00 0.N0 “.mofi H.000 0.00 0pxson H.N~H 0.¢0H o.¢~H m.m~H N.~¢H N.m- 0.00 0.~N~ m.mN— 0.NN~ “.Nn 0.o_~ 0.0m" N.0- meg afitmqm o.00~ 0.00H 0.o0H v.-~ $.0HH v.¢_~ ¢.mo~ ~.o0_ 0.m0~ _.v0~ 0.¢0~ 0._- o.mo 0.00m Hmm> m.m- 0.00H 0.oHH m.0- 0.0- ¢.w- 0.0- o.v- ~.0—~ 0.v- N.¢- w.—_~ w.o_~ o.0- mfiREHNumntu o.mHH m.0- v.¢HH m.NH~ N.0- ~.0~_ 0.0mm N.w- 0.0mu m.¢- ~.m- 0.0- w.00~ 0.00m xpom v.0HH ¢.HNH m.o- 0.0mm _.NN~ 0.0- 0.mN~ 0.vN~ 0.~N_ w.o- v.0- m.wo~ m.m- o.o0~ 000m H.0- 0.0H~ H.0~H v.m~H v.0- 0.w- 0._N— 0.0N~ 0.0uu 0.0- o.¢- m.oo~ o.0_~ 0.00u memos 0.000 0.000 o.H~H 0.000 w.__~ 0._- ~.md_ 0.v~_ 0.m_~ 0.00“ ~.c~_ v.0o~ v.0o_ o.mo~ :0: new . 0.53.ko 6ng 0.H0~ m.N0~ 0.HOH o.~0~ m.oo_ _.N0~ N.N0~ 0.00~ w.~0~ o.~o~ N.N0~ w.~0~ 0.N0— N.N0~ maUdUOum . . oumxdm m.~oH 0.NoH o.~o~ 0.00 0.oo~ v.~o_ ~.mo_ 0.mo~ m.mo_ 0.00" m.mo~ m.mo~ 0.00 N.~o~ omega” o.~o~ 0.NoH 0.000 N.Hc~ m._o~ _.No_ 0.No~ v.~o~ m.mo_ v.~oH 0.No_ N.No_ 0.00“ N.~o~ muosuo»0kr:wl $33 :83” . ----mm0_---. 0.00“ ~.moH 0.N0~ 0.00~ m.00~ v.N0~ m.m0~ ¢.mo~ N.N0~ m.~0~ v.m0~ ~.00~ N.00~ m.wo ma:3~HNMXX¢mmm m.m0~ 0.00~ m.00~ m.-~ ".00” ~.~0_ 0.m0~ v.mo_ ~.>0~ m.m0~ m.¢0~ o.oo m.00~ 0.m0~ mdomqfizvumas v.00 0.o0~ m.N~H N.0- 0.00M N.m0~ m.vw m.0w ¢.vm ~.00 m.00 0.00 o.oo w.~o mwmm m.Nofi N.00H 0.oo~ m.wo mwoo 0.vo_ 0.mo_ ~.¢0~ N.m0~ v.—o~ 0.m0~ 0.v0~ w.v0~. o.vo~ mdflvvcw mfiflm ~.No~ ~.oo v.00H 0.00 N.mo 0.No~ v.No_ ¢.No~ m.m0~ ~.mo~ 0.m0~ o.m0~ N.m0~ ~.mo~ mmmmum>mm 0.000 m.mo~ c.0og 0.000 N.oc~ m.mo~ _.mo~ 0.mo_ 0.00 o.oo~ m.00 o.~o_ o.Ho~ o.oo~ mvoow vegan . -20 333.30 m .00 012: m .00 N .00 m .00 0 .m0 N .00 0 .mo 0 .00 w .00 m .mo 0 .mo m .00 0 .00 33 mchon 0wmyv>m Na Na - Cu 0 w l .mwscq<. . Uofinwm wvmk.v_0. m v m N H mQZOHw 000“ v3: .3 H.493. l ~ WnNZCLI 33.. ‘ l.1:{|/I|7H/ \1‘1’ tll1lfl A 147 .0~ m~1~m~om 0.~0~ 0.$0~ 0.~0~ $.~0~ N.N0~ m.~0~ 0.m0~ N.N0~ m.~0~ 0.N0~ m.00~ m.00~ $.00 m.N0~ m000w 00.50 .920 >-mfihmn~ m .N0~ m .m0~ Nd: 0.00~ 0.0: 0.00 $.00~ N.~0~ 0.00 m .00 $.00~ m.00~ 0.00 m.0o m0flm 023300 ~.N0~ m .00~ 0.m0~ 0.00~ Ndo 0.$0~ 0.N0~ ~.~0~ $.$0~ 0.N0~ .N.m0~ $.~0~ 0.00 0.00 3mw.5m£pmm:m 0.00 ~.oo 0.m0 m .mo o.mo m .00 ~.0o o.mo 0.00 N.0o A“0.00 0.00 0.00 $.00~ vEcnlt pm 0000 005.00 o.m0~ ~.~0~ 0.N0~ $.m0~ 0.$0~ 0.m0~ 0.00~ m.$0~ $.00~ m.00~ 0.m0~ 0.N0~ 0.~0~ N00 moEmuomos. ‘ . ”a 0002.0 000000 N.No 0.00 ~.mo w.mo 0.~o 0.00 ~.mo N.mo 0.00 $.mw 0.00 0.00 Ndo m .00 mmfimpwmm.» CwNOMrm. N.0m~ 0.mm~ o.~m~ m d$~ $.0$~ N.$$~ N.m$~ m.wm~ N.0m~ $.Nm~ $.oN~ 0.mN~ 0.0: $.-~ midi ammouh 0.00 0.000 N.~0~ 0.00 0.00 0.00~ 0.00 0.00 0.00 ~.00 0.00 “.00 N.oo~ 0.00 mmifixomm> 090ch 0.00~ 0.N0~ 0.00 0.00 0.00 N.moH 0.00~ 0.00~ 0.00 0.00 0.00~ 0.000 0.00 0._0H 002:3.wocqmo 0.00~ 0.00~ 0.00 0.00 N.H0 0.00 N.~0 N.NHH o.0- 0.00~ 0.00~ H.0N_ 0.0HH .0.m- 000.0 .mofimuvwwm 0.00 0.N0 0.00 N.mw H.00 0.Nm ~.H0 m.0_H 0.0H_. 0.NHH 0.0““ 0.000 0.000 0.00 ammuw.mgufirm 0.00~ 0.00~ 0.N0~ 0.00 0.00 0.00 0.N0~ N.0~H m.o~H 0.-_ 0.00~ 0.00~ N.0o~ H.000 mquxmmo>ll 0cm mfiah ¢ owMV Ho 0 o o o o o o o o o o I. 00 0 00 N 00 m 00 N 00 o 00 0 00 $ 00 ~ m0 w wo N 2: 0 mo 0.00~ 08:00.20 0.35 0wmum>m m~ N~ - {l 00.50“de 0~ 0 w 0 o m V m. N 03.80 v0.53 $ ~ . mQSOpw 000% 148 CHAPTER VII COMPARISONS OF THE INDEXES CALCULATED BY USING DIFFERENT PRICES AND WEIGHTS, AND THE SAME PRICES BUT DIFFERENT WEIGHTS, FOR ALL FOODS Introduction Several comparisons, between the B. L. S. series and the M. S. U. Consumer Panel price series, have been made in Chapters III and IV. This chapter deals primarily with the construction of indexes, using different prices and weights, and using the same prices but different weights. Comparisons were made in order to: (1) evaluate the differences between the three food price indexes--the M. S. U. Consumer Panel food price index (based on the panel prices for Lansing), the B. L. S. food price index (based on the store sample for Detroit), and the N. I. C. B. index (based on the store sample for Lansing); and (2) test the effects of using the B. L. S. weights for Detroit and M. S. U. Consumer Panel weights for Lansing (both based on the panel prices) in constructing the food price index for Lansing. Comparisons of the M. S. U. Consumer Panel Food Price Index with the B. L. S. Index for Detroit, and the N. I. C. B. Index for Lansing * Comparison of the M. S. U. Consumer I'Panel retail food price index and the Detroit B. L. 5. retail food price index. --Figure 31 shows the movement of the ”all food at home”. index of the M. S. U. Consumer 149 Panel for Lansing and the B. L. S. for Detroit. The Detroit B. L. S. food price indexes (1947-49 = 100) were converted to the same base period as was used for the M. S. U. Consumer Panel (1955-57 = 100) in order to facilitate comparisons from time to time. These two indexes moved together fairly closely, with differences in price levels being due to the item specifications and the levels of the base period prices and the actual prices in the given periods. The correlations between the two indexes was . 82, which is significant at the.l percent level. Both indexes exhibit an upward trend. However, the Detroit B. L. S. index fluctuated over a longer duration than the Lansing M. S. U. Consumer Panel index. In comparing the annual average indexes, the B. L. S. series increased . 8 percent from 1955 to 1956, 2. 5 percent from 1956 to 1957, and 3. 3 percent from 1957 to 1958, while the M. 51 U. Consumer Panel series increased 1 percent, 1. 7 percent, and 1.7 percent, respectively. The panel index showed more'period- to-period variability. Presumably this variation in the panel index reflected more of the effects of the prices of "specials" and, to some extent, changes in the quality of some of the products purchased by the panel families from one period to another. Comparison of the M. S. U. Consumer Panel retail food price index and the N. I. C. B. retail food price index, for Lansing.—- The N. I. C: B. food price index was calculated four times a year; in March, June, September and December. These months correspond closely to the 3rd, 6th, 9th and 13th periods, of the panel series, 150 32 33 on: 0.0.2 ~____~____~_--0_____-—~H-~_—_--l—.~_--~d-_ Oh I 8 I 00 u \01 .:.. ....2132.11.01.a...l..§..l...o..:.. For...o...o...o...o.lo Sfihnsnz I .OIOI ul 000m -< how Nova—H ouwum 68h ~«3QM ~dfldx~ ulna-8 41%.! o- on: 00.2 0mg nn0~ ___.__~___d_______4_~a~_.______fi~_1.qd:.401fifiad~ oh 1 on .I 00 ‘c‘: ‘.-.I.‘.' P’I:l.'.‘ \.\. .-.. ‘ ‘.\. ... o0.— v.¢.o.0.l. \.\.\.. o.‘.!.£.\ .<. «.-:.... 811.0000: 3 303.68 I 62.00-32 .3058 you .3ng3: v80 :33. .0....» o: sinus“: .u «aim-hhmzzfiun .no a” 52.3 new 7:: hols-sou .:.m... 23 van 3033 now 6.4.0 33 .3253 ~ I. uguowuuv 95.: 03393.0 .noxovnu coca coon :30... any we scanning .:.. .lwh 151 «nod snag onod «nag nd a o n mg a o n ad a o n nd a o n _ _ . _ _ fl . fl _ A _ _ _ _ _ _ no: and» «nah an: ova uaom «nan aux con anon cash an: 009 uaom scan us: IIIJ nL ogflnénfi .28: u. .38 :4 you .263 I 00.7mm 30h meld-OI ..0—aim king 41%.! I1 ’ 1+l‘ d a} I |. o o ' o h I, b|ol ‘ o 0 v . . .r . . .... . . t. . oSHS$nE 3 vouu>\n8 .8122 I .ma«.u¢q you nova» «can; coon duququ ...o.~.z 321...-32 v.83: denim aqua-coo .:.».x on» can vane: ouaououcou anauudaunn guacaunz cnu coo3uoa .ooagldo uaououuuv «can: vuudusoado .and-and new nouovaw oouun team ~«Quou on» we conga-aloe .2 .u: . cm on om cod odd and on co cod Gag 152 respectively; The Mo 8. U0 Consumer Panel index and the No 1., Ct B1 index for Lansmg were compared for these periods. The No L. C, B, index included both food eaten at home and away from home, while the Mo So U. Consumer Panel index was for food eaten at home onlyo A graphic comparison of the food price indexes for Lansing between the panel and the Ni Io C. B. series is shown in Figure 320, The panel food price index moved parallel to the No I. C. B. food price index despite varying specifications of the food items, The correlation between the two indexes was a 810 This is significant at the 1 percent level, Although the two indexes moved parallel to each other; there were wider fluctuations in the N... L. C0 B0 series than in the panel series: In comparing the annual average indexes, the N019 Ct Bo series went. up 13 3 percent from 1955 to 1956., 20 4 percent from 1956 to 195.7, and 5.4 3 percent from 1957 to 1958, while the panel series went: up 1 percent, 1;. 7 percent, and 1° 7 percent: respectively, The wider fluctuation in the N; 1° C9 B0 index probably resulted from the combined effects of the greater relative importance. of meats and fresh fruits and vegetables in the N, L Cu. B. series than in the Mo So Uc Consumer Panel series, The differences between the store sample prices and panel sample prices suggest another explanations 153 Comparison of the Indexes, Based on the Same Prices, but Calculated by Using Different Weights It seems desirable to evaluate the differences between using Detroit B. L. S. weights with the panel prices and the M. S. U. Consumer Panel weights with panel prices to construct a Lansing retail food price index. The B. L. S. weights represent the average family expenditures for foods of Detroit urban wage-earner and clerical worker families in the year 1952. These expenditure weights are currently used by the B. L. S. to calculate the retail food price index for Detroit and, at the same time, to combine Detroit with other cities to form the U. S. city average index. As we know, the proportion in which the consumer's dollar is shared among different kinds of foods changes over time. Even though the division of family expenditures on food does not change rapidly, small shifts of the relative impor- tance of different groups of foods will affect the total food index. Considering the overall changes in eating habits that have taken place and the greater variety of foods that have been offered, as income increased from 1952 to 1957, it is reasonable to assume that the division of the food dollar spent by the Detroit families has also been changed from 1952. to 1957. The division of expenditures for foods in different cities shows the effects of regional food habits. In the southern part of this country, in a city like Birmingham, Alabama, cereal and bakery products 154 presumably take a larger share of the food dollar than in the northern cities like Detroit and Lansing. The different levels of income between the North and South suggest one explanation; and Southern customs and eating habits indicate another. It would be reasonable to assume that the relative importances of the major components in family spending patterns for foods are about the same for both Detroit and Lansing families. In a broad sense, United States cities have spending patterns which resemble one another rather closely. The known expenditure patterns of neighboring cities have been used a number of occasions in the past, both by the B. L. S. and the N. I. C. B. , for estimating the expenditure patterns for unsurveyed cities. These were done on the assumption that graphic proximity is an adequate basis for estimation (other factors such as income, climate, industrialization, family size are also correlated with spending patterns). An effort to explore some of the differences between the B. L. S. expenditure patterns of foods, for the U. S. city average, and the M. S. U. Consumer Panel expenditure patterns, for Lansing, has been made in Chapter V, Table 9. The following graphic comparisons were made for two purposes. The first is to analyze the differences between the Lansing food price indexes using Detroit B. L. S. weights and panel prices, and using panel weights and panel prices. The second is to evaluate the bias of the B. L. S. food price index for Detroit, assuming that the Detroit families distributed their 1955-57 food expenditures in the same manner as the Lansing families did in the 155 “mud n d onad “n3 122.22.2122 ail:SIZES—1.234.... 3:3... .4 «as we .3100..— Id 3033 won 3303 ends» we .0.-dunes again: ..:».- II .328 :33 a: 3 33.8 3.8 ulsaoo .96... .3. 83 3 83r— tnansvouaai 1:8 “...-:80 .:.»... us... 33:58 .33 as on nu ca x or ...:.. / \\ I; .0 ....o... .... K .... 3:33 .1 n2 ..:...:... .4 9:33 you 3.3:. 833.55. 2-22 13.. all-lo .:.»... can .33: levee 05 3 .03.: decay hols-coo I o: .:.»... a... :3 .4 .328 «3.3033325 15: .3338 .:.»... 9:3 138-38 .83 1 n: l o2 Azalnunnanusau sauna-... new 33!- ..ousn unis-.50 .:.“... van . 33qu you 34:!- .m..—.n new: voulaoudo ..:.—.5 03.3 coca 23 mo nasal—loo .nn 9; 156 1955-57 period and that the Detroit food prices increased at the same rate as the Lansing food prices from 1955 to 1958. The two indexes, as shown in Figure 33, were both cal- culated by using Lansing prices obtained from the M. S. U. Consumer Panel. The first, represented by the dashed line, was calculated by using the B. L. S. weights for Detroit; the second, represented by the dotted line, was calculated by using the panel weights for Lansing. \ In the beginning of 1955, as shown in the figure, the two food price indexes converged. This was mainly due to low prices in dairy pro- ducts and the high prices in fresh fruits and vegetables, and egge. The dairy products were relatively more important in the panel weights for Lansing than in the B. L. S. weights for Detroit; and the fresh fruits and vegetables and eggs were relatively less important in the panel weights for Lansing than in the B. L.S. weights for Detroit. During this period, the Lansing food price index would be the same whether the Detroit B. L. S. weights or Lansing panel weights were used. The similarity resulted mainly from the fact that the weight differences were offset by the different rate of the price changes between food items. From the 4th period to the 9th period of 1955, the prices of fresh fruits and vegetables, and eggs continued to increase sharply while the prices of dairy products levelled off. During this period, the dashed line (representing the index calculated by the Detroit B. L. S. weights) moved above the dotted line (representing the index 157 calculated by the panel weights). This was due to the high prices of fresh fruits and vegetables, and eggs. The higher prices were accompanied by the low prices of fish in this period, but fish con- stituted a relatively less important component in the Detroit B. L. S. weights than in the panel weight. In this period, the Lansing food price index would overestimate the cost of foods for Lansing families, if the Detroit B. L. S. weights were used in constructing the index. At the same time, the B. L. S. food price index for Detroit would have biased upward the estimate of the food for Detroit families, if the Detroit 1955-57 family spending patterns were the same as the panel 1955-57 family spending patterns. From the 9th period of 1955 through the 13th period of 1958, the index represented by the dotted line gradually moved above the index represented by the dashed line. The continued increase in prices of dairy products from the latter part of 1955 to the early part of 1958, accompanied by the sharply lower prices of eggs, were mainly responsible for this difference. The drop in the prices of dairy products in 1958 and the sharply decreased prices in the fresh fruits and beverages compensated for each other in the two indexes. Thus, the index which was calculated by the panel weights was still above the index which was calculated by the Detroit B. L. S. weights despite a slight decline in dairy product prices in 1958. From the 9th period of 1955 through the 13th period of 1958 (except the 7th period of 1956), the Lansing food price index would underestimate 158 the cost of foods for Lansmg families, if it were based on the B. L. S weights for Detroit. At the same time, the B. L.S. food price index would have a downward bias in estimating the food cost for Detroit families, if their spending patterns for food in the period of 1955-57 were the same as the panel families in the same period and the rate of increase or decrease in the food prices in Detroit were similar to the price changes of the food items in Lansing. When the index, calculated using the panel weights, moved below the index which was calculated using the Detroit B. L. S. weights, it was an indication either that there were price reductions in foods of. which Lansing families buy relatively more or that the price increases were in foods of which Lansing families buy relatively less. In this period, the Lansing families would spend less money on food items than the Detroit families if the spending patterns for foods of Detroit families remained the same as they were in 1952. When the index, calculated using the panel weights moved above the index which was calculated using the Detroit B. L. S. weights, it was an indication either that there were price reductions in food of which Lansing families buy relatively less or that the price increases were in foods of which Lansing families buy relatively more. The Lansing families would spend more money on food items in this period than the Detroit families if the spending patterns of Detroit families for foods remained the same as they were in 1952,. It seems unlikely that the Detroit families spent the same amount of money on each group of food items in the period of 159 1955-57 as the same as they did in 1952. It is very likely that the Detroit families have similar spending patterns in the period of 1955-57 to the Lansing panel families in the same period. In this case, if the Detroit family spending patterns in 1955-57 were the same as the Lansing families, the B. L. S. index for Detroit would have under- estimated the cost of foods for Detroit families from the latter part of 1955 through the end of 1958 (except the 7th period of 1956). Like- wise, the B. L. S. food price index for Detroit would have overestimuedj but very slightly the cost of foods for Detroit families from April to September of 1955. (See Figure 33.) In comparing the annual averages between the two indexes, calculated by using different weights, the index with panel weights stood at 99. 1 in 1955, 100. l in 1957, and 103. 5 in 1958, while the index with the Detroit B. L. S. weights stood at 98. 9, 98. 5, 97. 6, and 100. 2 respectively. This indicates that the index with panel weights was . 2 percent higher in 1955, l. 6 percent higher in 1956, 4. 2 percent higher in 1957, and 3. 3 percent higher in 1958 than the index with the Detroit B. L. S. weights. 160 CHAPTER VIII THE STRENGTH AND WEAKNESSES OF THE M. S. U. CONSUMER PANEL RETAIL FOOD PRICE INDEX IN COMPARISON WITH THE B. L. S. RETAIL FOOD PRICE INDEX This chapter undertakes a comparison of the strength and weaknesses of the M. S. U. Consumer Panel retail food price index and of the B. L. S. retail food price index. The problem of constructing an index number is as much one of economic theory as it is of statistical technique. Most of the authors who have discussed index numbers have laid a strong emphasis on the problem of sampling and the choice of formulas. King says: "The problem of index number arises entirely from the fact that index numbers are problems of sampling, nothing more and nothing less. If complete data are used, it is easy enough to give the formula that will answer any definite question--hence there is no problem involved. " Fisher gives over 150 different formulas for the construction of index numbers. 2 He recognizes, however, the other sources of errors in index numbers: choice of items (if a price index is in question), number of items included, and errors in the price quotations. Any one l Willford I. King, Index Number Elucidate (New York: Longmans, Green and Co., 1930), p. 49. 2 Irving Fisher, The Makian Index Numbers (New York: Houghton Mifflin Co. , 1927). 161 of these, he admits, may be more important than the choice of formula. But at the present time, he states, "The chief source of error is the formula. "3 This type of error could, in his opinion, be eliminated simply by using one of the many good formulas which he presents. Each of these other sources of errors “offers a field of study which has scarcely been touched. " The sampling procedures and the selection of formulas for the index number construction have been discussed by many index number writers. It has been recognized that the practical necessity of building index numbers upon incomplete data does in fact make the sample as important as the choice of formula, in any index number construction. The consensus of opinion to date is that we would have many technical ‘problems in constructing an index number even if we had complete data and appropriate formulas. With this in mind, therefore, an appraisal of the strengths and weaknesses of the panel food price index in comparison with the B. L. S. retail food price index, is not limited in the areas of sampling and formulas. The following brief discussion focuses on the differences between the two series of food price indexes, referring particularly to: (l) the methods of obtaining the spending patterns from a representative sample of families; (2) the techniques of choosing the items which are priced in the index; (3) the method of collecting prices; (4) the index formulas 31bid., p. 342. 41bid., p. 331. 162 used; and (5) the weights used in index construction. The methods of obtaining the sfinding patterns from a representative sample of families. --The B. L. S. family spending patterns, which formed the B. L. S. index weights, were determined by the 1950 survey. A complete record of the kinds, qualities, and amounts of foods was obtained by the interviewers who visited and inter- viewed those sample families which were selected by the B. L. S. to represent all city wage earners and clerical workers. The M. S. U. Consumer Panel family spending patterns, which formed the panel index weights, were determined by taking an average of the reporting diaries for the 3-year period, 1955-57. Comparison of these two methods of collecting information from consumers shows that the panel method offers more strength than the B. L. S. direct survey in the following aspects: In the first place, the panel method is more sensitive to change in individual behavior from time to time. A 3-year average of weekly purchases, which was used to determine the spending patterns of the panel families, includes the changes in the consumption behavior from week to week over the entire 3-year period. The B. L. 5. direct interviews, which were used to determine the spending patterns of the B. L. S. sample families, can only provide the information of kinds, qualities, and amount of foods bought at a particular time in the year of 1950. This information, which was provided by the interviewers, covered only the food purchases in a 7-day period. The information obtained may 163 be biased by the seasonal purchases of the families. Direct interviews can be done quickly for a specified period and the results become available in a short time. They cannot include information on changes in consumption behavior over time. Secondly, diary reports from panel families furnish more accurate information on food purchases than data which were obtained from interviews. It seems likely that those who are interviewed cannot recall their purchases as accurately as those who keep a diary record for their food purchases from week to week. On the other hand, the panel method has the following weaknesses in comparison with the recall method: In the first place, the panel members may become conditioned to the items in the re- cording diary and may begin to exhibit a different kind of buying behavior than they would if they were not members of the panel. In this aspect, the direct survey may also bias the information collected by interviewers. Those who are interviewed are often affected by the interviews so that they usually become different from those who are not interviewed. The consequent loss of representativeness is obvious. Second, as time goes on, some panel members drop out and are replaced by new members. This may introduce an initial bias. In order to guard against this bias and maintain a representative sample of the universe, the M. S. U. Consumer Panel conducted three sample censuses over the 8-year period. Therefore, replacement members in the panel do not seriously affect the representativeness 164 of sample. Thirdly, the consumer reporting panel is an expensive operation, particularly on a national scale. It also requires constant and expert administration. These may be a weakness of the panel method of collecting information, but they are not a weakness of the panel retail food price index itself, since the panel data were already collected for other purposes before the construction of the panel index was initiated. The techniques of choosing the items which are priced in the index.—- The techniques of choosing food items, which are used in the panel food price index, are the same as those used in the B. L. S. food price index. The selection of foods includes: items that are relatively important in family spending, items that are representative of price change for groups of related products, items that have distinctive pricemovements of their own, and, in some cases, several qualities of items priced to represent a single item. However, the panel index priced many more food items than the B. L. S. food price index. (For comparisons of the food items included in the panel series and the B. L.S. series, see Table 9, Chapter V). If these two samples are equally well representative of their population, then the more items priced in the index the smaller the error in measuring the average price change of all foods will be. The method of collecting prices. —-If the index is specifically designed for the purpose of measuring changes in cost of foods due to changes in prices, the strength of using panel methods to collect prices 165 in comparison with the B. L. 8. store sample prices is obvious. The panel price series is the weighted average price. The price of each item is the total expenditure divided by the total quantity, bought by all panel families for each week. The B. L. S. food prices are collected from the food store only once a month. They are collected during the first three days of the week which includes the 15th of the month. Itis fairly well accepted that most of the housewives buy their foods during the week end in order to take advantage of the week- end specials and discounts. The B. L. S. store prices are collected in the first three days of the week when only a small proportion of the total amount of foods are bought and few specials and discounts are offered. In this case, the panel price reflects the actual prices paid by the consumers for each food item purchased during the week, while the B. L. S. store sample price reflects only the prices charged by the stores in the specified time of each month. The food price index, which was calculated using panel prices, more accurately reflects the prices of food paid by the consumers than the index which was calculated using the B. L. S. store sample prices. On the other hand, the panel prices are not as explicit as the B. L. 5. store sample prices. The panel prices which are paid by the panel families are for items of varying sizes and qualities. The B. L. S. store sample prices are for particular specifications of each given item. The price changes from one period of time to another are not due to quantity and quality differences, but only reflect changes in prices. 166 Therefore, the food price index which was calculated using panel prices not only reflects the price changes from one period to another but also reflects the differences in quantities and qualities purchased by the panel families from week to week. The index formulas used. --The B. L. S. has adopted a chain index for its revised index with each link calculated by the Laspeyres' formula. The chain index simplifies the linking procedure when the weights are revised. The Laspeyres' formula was also used in combining prices in the panel index. Because the weights used in the panel index represent the average panel family expenditure patterns of the three years 1955-57, the panel index did not require any weight revision or linking. In this case, the panel index is an aggregative index constructed on a fixed base; commensurability exists between any two periods of time. The weights ascribed to each component of such an index defined the relative magnitude of each such com- ponent in relation to all others and to their totality. Each component of the index is made comparable to all others. Similitude, which is the essence of all measurement, however, is not present in the case of chain indexes. When such indexes are constructed, the relative magnitude of the different components is not defined uniquely over the entire span of the time sequence. An index number composed of several different sets of weights linked together has no operational significance since a set of magnitudes derived on a basis of fixed weights during one period of 167 time is not commensurate with another set of magnitudes derived from a differing weighting pattern in another period of time. The weight-fixed price index has a theoretical advantage if the index, for example, the panel food price index, is only applied over a short period. The panel food price index for total food was only constructed for the 1955 to 1958 period, because of the lack of data. However, the weights used in the panel food price index were based on the average of the 3-year (1955-57) expenditure patterns of the panel families. Therefore, the panel food price index did not require revision of the weights or linking. If the panel index continues over a longer period of time, the index would have to be chained if the weights were to be revised. In this case, the panel index would become a chain index like the B. L. S. index. Thus, the panel index would have no theoretical advantage. Weights used in the index construction. --The "market basket" is the basic framework of the index. It is based on family spending patterns. But the spending patterns change over time. It is because of the changes in people's buying habits that the index weights have to be revised. The revision of weights reflects this change and brings the index up to date in terms of current buying. The new "market basket" of the B. L. S. revised index was based on the survey of consumer spending patterns in 1960. The 1950 family spending patterns were adjusted for changes between 1950 and 1952. This provided the basis for the new "market basket" of 168 of the year 1952. The panel index weights were based on a 3-year (1955-57) average of expenditures of the panel families. The resulting index of the B. L. 5. series is a measure of the effect of price changes on the cost of food in the 1955-57 "market basket. " As time goes on, the changes of the consumption patterns will cause an increasing lack of representativeness of the B. L. S. family spending patterns and more errors will be introduced in the B. L. S. food price index. It is evident that the panel food price index which was calculated in terms of the 1955-57 ”market basket'l is less biased by this kind of error since the time period was short. In summary, the panel index has both strength and weakness. The panel index provides an accurate measure of the costs of food since the spending patterns represented the actual purchases of the panel families and prices are those actually paid by the panel families. The panel index is less biased because the weights used were the 1955-57 spending patterns of the panel families. On the other hand, the panel price changes may be affected by the composition of the quality and quantity of food items, purchased by the panel families from one period to another. In selecting a preference for one index over the other, one would need to choose the type of error which is least. harmful, the error caused by changes in qualities and quantities in the panel data or the error caused by selecting prices three days of the month when very little food is bought, selecting specific qualities of a product which may represent only a small proportion of 169 that actually bought, selecting a method of distribution, the retail store, to represent all methods of distribution, and selecting specific sizes and types of containers which may represent only a small proportion of all sizes and types for a product, as is done in the B. L. S. data. 170 CHAPTER IX SUMMARY AND CONCLUSIONS The principal objective of this study was to compile an M. S. U. Consumer Panel retail food price index. The index is a statistical measure of changes in the retail prices of foods bought by Lansing urban families. A retail food price index is simply a part of a consumer price index. Consumer indexes are very important and widely used in demand analysis. Foods account for a major part of total family spending and constitute a very important component in the consumer price index. Because food prices change very frequently and because some food prices are seasonal, food forms a major problem in making up a consumer price index. It has been recognized that the problems of constructing a retail food price index are many and difficult. But not all of the problems are of equal importance nor are they always independent of one another. The problem which we face is that of providing an accurate index for measuring the price movement of retail foods. The errors introduced into the index numbers are from different sources. Most of the index number writers recognize, however, five main sources of error in a price index number: errors in price quotations, choice of commodities, number of commodities included, selection of sample, and choice of formula. 17.1 The source of data, for constructing the retail food price index for Lansing, was the M. S. U. Consumer Panel. Weekly obser ., vations on prices, quantities, and total expenditures were available for meats, poultry, fish, dairy products, eggs, and fats and oils from 1951 to 1958, for fruits and vegetables from 1953 to 1958, and for a complete price series, including every item in the diary, from 1955 to 1958. The various series of prices were computed by dividing the total expenditures by the total quantities purchased by the entire panel. It is quite clear that the M. S. U. Consumer Panel was not specifically designed to collect data for the purpose of constructing a food price index. The panel prices were those actually paid by the entire panel families. They were the weighted average prices for items of varying sizes and qualities. Therefore, the first problem encountered was that of examining the reliability of the panel data for constructing the retail food price index. As noted before, the type of price quotation is often one of the main sources of errors in food price indexes. Three problems exist: (1) whether the panel prices accurately reflect price changes in the retail stores; (2.} whether the panel data are sufficiently accurate; and (3) whether the. quality of the products in the panel is sufficiently consistent over time. During the life of the panel, the food items were some- times changed in order to get better information or more accurate reporting. In addition, tabulations of prices on food items were 172 changed on different occasions. Some. data were adjusted in order that every item in the index would represent a relatively uniform quality of food from the beginning to the end. In some cases, sever al different qualities of items, which were combined in the panel data, were separated into individual qualities, or in other cases, several individual qualities were combined to represent a single item, for calculating the index. The original panel data and tabulation methods did not insure the comparability of pricing items from one period of time to another. In order to determine the comparability of the panel prices from period to period, several comparisons were made. The panel prices for Lansing were compared with prices for similar products which were collected from food stores by the Bureau of Labor Statistics for constructing their U. S. city average index, and their index for Detroit. Firstly, comparisons were made between the panel prices for Lansing and the B. L. S. prices for the U. S. city average. 'Thése comparisons were: (1) the levels of the two price series; (2) the move.-- ments of the two price series; (3) the closeness of the two price series; and (4}- the magnitude of price changes of the two price series. Dnect comparisons of the price levels are difficult because there is con- siderable difference in qualities between the B. L. S. quoted items and the items actually purchased by the panel families. Comparisons of the level of prices between the two series not only indicate the dif'ierenc e. H3 in prices but also reflect the difference in qualities. More important was the comparison of the price movement over time for similar items. These two comparisons, price level and price movement, were made by using index numbers and graphs. Regression and correlation analyses were used for comparisons of closeness and fine magnitude of price changes between the two price series. In order to compare both levels and changes between. 'he B. L. S. price series for the U. 5. city average and the panel series for Lansing, the i'first index" was compiled. This term, "first index, -‘ is used in order to distinguish it from the other indexes in this disser tation. This index was compiled using the same weight and base as the B. L. S. used, but uSing the M. S. U. Consumer Panel prices in the given periods. The resulting index and the B. L. S. index pr ovide a means of comparison of the price levels and price changes between the panel price series and the B. L. 5. price series. The index was calculated for each group of products and plotted on graphs in order to facilitate comparisons with the B. L. S. index. (See Figures 1~»-7. f. Comparisons of the two series indicate that the B. L. S. quoted prices for the specified products were higher than the panel price in each group of Similar products. The higher level of prices in the B. L. S. series compared with those in the panel series was mainly due to the fact that the B. L. S. specified qualities were usually above those qualities actually bought by the bulk of. consumer 5. Comparisons of the price changes between the two price 174 series show that both the B. L. S. and the panel series followed similar patterns of price movements in each group of the products. However, the rate of price change in the panel series relative to the rate of change in the B. L. S. series was not exactly the same. Nevertheless, the two price series moved parallel to each other despite some differences in qualities between the two samples. This comparison of the panel prices with the B. L. S. prices, which were specified for homogeneous qualities of products, provides some evidences of the reliability of the panel data for constructing price indexes. In determining the closeness and the magnitude of price changes between the B. L. S. series and the panel series, a series of correlation coefficients and regression coefficients were calculated. These coefficients were calculated using actual prices, and each pair of group indexes, for the two series. In calculating these coefficients using actual prices, the equation was fitted by traditional least square regression methods. The mathematical form of the equation was as follows: y = a + bx where x : prices of the B. L. S. products for the U. S. City average y : prices of the M. S. U. Consumer Panel products. The two groups, cereal and bakery products, and meats, were calculated using the actual prices. The reason for choosing these 175 two groups is that the prices of cereal and bakery products did not follow the B. L. S. movement very closely, while meats did follow each other very closely, as shown in the graphs. The correlation coefficients indicated that only four of the nine items of the ”cereal and bakery products" group were found to be significantly correlated with each other. These four products were rolled oats, corn flakes, bread, and soda crackers. When dealing with meats, all meat items were found to be significantly correlated with each other. When applying regression analysis, the question arises as to whether or not the trend should be removed before the regression analysis is commenced. If a trend is present in the price movement, this may dominate the short-term fluctuation. Trends may give rise to spurious regression. As shown in Figures 1-7 (also Figures 8-14), there was no food group showing a clear trend, except the "cereal and bakery products" group, which showed a slight upward trend in the B. L. 5. series. The results of regression analysis for the cereals and bakery products indicate that only the panel prices of corn flakes varied more widely than in the B. L. S. series. The negative re- gression coefficients show that the prices of wheat flour and corn meal in the B. L. S. and the panel series changed in the opposite directions. This was due primarily to the effect of the greater frequency of the price fluctuations in the panel series. The B. L. S. prices gradually increased while the panel prices fluctuated up and down from one 1'76 period to another. When dealing with meat items, the regression coefficients indicate that only the prices of rib roast and lunch meat in the panel series fluctuated more widely than in the B. L. 5. series. The panel prices for other meat items fluctuated less than in the B. L. S. series. In determining the closeness and the magnitude of price changes of each group of products, between the B. L. S. series and the panel series, the indexes of each group of products were used. The correlation coefficients and regression coefficients were calculated for each group of foods and the all foods group. The estimating equation was: y = a + bx where x : index of B. L. S. groups of products for the U. 8. city average y = index calculated by using B. L. S. weights and base, but using the M. S. U. Consumer Panel prices in the given periods. The correlation coefficients between the two price series, as measured by the indexes of each food group, were all highly signi- ficant. These results indicated that the prices of each pair of food groups, between the two series, moved very closely with each other. The regression coefficients, as calculated for each pair of the groups of products, indicate that only the prices of the “fruits and vegetables" group changed more widely in the panel series than in the B. L. S. series. The prices of other groups changed less in the 177 panel series than in the B. L. S. Series. The correlation coefficients and regression coefficients calculated using indexes of food groups were as follows: Correlation Regression coefficients coefficients (1') (b) Meats, poultry and fish . 95 . 95 Dairy products . 43 . 50 Fruits and vegetables . 91 l. 25 Cereal and bakery products . 48 . 24 Other food at home . 48 . 74 All food at home . 91 . 66 The above results (graphic comparisons and least squares regression and correlation analysis) indicate that the panel prices for Lansing displayed a close relationship to the B. L. S. prices for the U. S. city average. The price levels and changes were reflected in both series of prices. However, the panel prices showed more variability and lower prices of the products than the B. L. S. prices for the U. S. city average. This greater frequency of variations and lower prices of products reflected the effect of "specials, " and of changes in composition of the purchases of panel families, from one period to another. In order to compare price changes between the B. L. S. price series for Detroit and the panel price series for Lansing, a "second index" was compiled. Because neither the price of each item in the base period nor the index of each item at any later date was available, the B. L. S. price base used for the Detroit index could not be derived from the available data. Therefore, the second index was 178 designed only for comparison of price changes between the B. L. S. series for Detroit and the panel series for Lansing. This index was calculated using the B. L. S. "relative importance of value weights‘E (in the retail food price index for Detroit), the M. S. U. Consumer Panel average price (1955-57) as base, and the M. S. U. Consumer Panel prices in the given periods. Comparisons of the price changes between the two series, after plotting the indexes on graphs, indicate that the changes in prices between the B. L. S. for Detroit and the panel for Lansing were about the same as the B. L. S. for the U. S. city average and the panel for Lansing. Each pair of food groups, of the two price series, followed a very similar pattern and moved up and down together fairly closely. In calculating the correlation coefficient and the regression coefficient using actual prices, the estimating equation was as follows: y = a + bx where x 2 prices of the B. L. S. products for Detroit y = prices of the M. S. U. Consumer Panel products. The correlation coefficients indicated that the relationships between the panel series for Lansing and. the B. L. S. series for Detroit were the same as the relationship between the panel series for Lansing and the B. L. S. series for the U. S. city average. The regression coefficients of actual prices for cereals and bakery products show that the relationship between the panel for Lansing and t the panel for regression c rib roast, h panel series products in series. Ii coefficients u was as follow of value weig Price (1955- given period ___, *_‘ , for each ngi cOefiiciems B°L'S~ ser; Primarily h B: L. 8' pri\ C I79 Lansing and the B. L. S. for Detroit were the same as those between the panel for Lansing and the B. L. S. for the U. S. city average. The regression coefficients for meat products indicate that the prices of rib roast, ham, lamb, and lunch meat changed more widely in the panel series than in the B. L. S. series. The prices for other meat products in the panel fluctuated with less amplitudes than the B. L. S. series. In calculating the correlation coefficients and regression coefficients using the index of each food group, the estimating equation was as follows: y = a + bx where x 2 index of the B. L. S. groups of products for Detroit y - index calculated using: the B. L. S. "relative importance of value weights" for Detroit, the M. S. U. Consumer Panel average price (1955-57) as base, and the M. S. U. Consumer Panel prices in the given periods. The correlation coefficients between the two price series for each group were also highly significant. However, the correlation coefficients were higher than those between the panel series and the B. L. S. series for the U. S. city average. These results were due primarily to the fact that the panel prices moved more closely with the B. L. 8. prices for Detroit than with those for the U. S. city average. The regression coefficients show that the relationships of price changes between the panel series and the B. L. S. series for Detroit were the B. L. 5= only the P51T widely than other than t series than and regress as follows: Meats, poultr Dairy product Fruits and ye Cereal and be Other food at All food at. ho lz‘ Linsing faml : . expenditui-e : Centage distrr annual aV’era indes The , the Panel fan on the "meat 180 Detroit were similar to the relationship between the panel series and the B. L. S. series for the U. S. city average. Among the food groups, only the panel prices of the ”fruits and vegetables" group changed more widely than the B. L. S. series. The price indexes of the food groups, other than the "fruits and vegetables" group, changed less in the panel series than in the B. L. S. series for Detroit. The correlation coefficients and regression coefficients that were calculated using the indexes were as follows: Correlation Regression coefficients coefficients (1') (b) Meats, poultry and fish . 95 . 91 Dairy products . 60 . 82. Fruits and vegetables . 91 1. 15 Cereal and bakery products . 49 . 32 Other food at home . 60 . 76 All food at home . 91 . ‘76 In order to determine the food expenditure patterns of Lansing families, for use as expenditure weights in the index, the panel family expenditure distribution was measured and compared with the expenditure pattern for foods for families in other samples. The per- centage distribution of food expenditures of the 3—year (1955-57) annual average formed the ”value weights" of the panel food price index. The actual distribution of expenditures (1955-57) showed that the panel families spent 30. 66 percent of their total food expenditures on the "meat, poultry and fish" group; 20. 34 percent on the "other fOOd at home" group; 20. 03 percent on the "dairy products" group; 16. 93 PEr on the C‘ igures i families to buy 1hr average 1 period of more mom and less on amount of rr than they dlC of foods that University C sample for . Board 5am; Sample for ' diffemnces importance families m . 1952. fOr the Lansing. a1: 181 16. 93 percent on the “fruits and vegetables" group; and 12. 03 percent on the ”cereal and bakery productsH group. The relative importance figures for other than the base period (1955-57) indicated how Lansing families would distribute their expenditures on foods if they continued to buy the same kinds and amounts of foods they purchased on the average in 1955-57. The relative importance figures in the 13th period of 1958 showed that the Lansing families would have to spend more money on “meat, poultry and fish" and “fruits and vegetables" and less on "dairy products" and ”other foods at home” and the same amount of money on "cereal and bakery products" in December, 1958 than they did in 1955-57, if they were to buy the same kinds and amounts of foods that they purchased in 1955-57. (See Table 9, Chapter V.) Four different samples of families--the Michigan State University Consumer Panel for Lansing, the Bureau of Labor Statistics sample for the U. S. city average, the National Industrial Conference Board sample for Lansing, and the Agricultural Marketing Service sample for the United States—-were used in order to compare the: differences between various expenditure patterns. The relative importance figures represented the spending patterns for the sample families in the periods of 1955-57 for the M. S. U. Consumer Panel, 1952 for the B. L. S. U. 5. city average, 1953 for the N. I. C. B. for Lansing, and 1955 for the A. M. S. for the U. S. Comparisons of these relative importance figures between the different samples showed that panel families distributed about the same proportion of their total 182 food expenditures on the "cereal and bakery products” group and the ”meat, poultry and fish" group as the B. L. S. and N. I. C. B. samples of families. But they spent 10 percent less and 6. 4 percent more on the "cereal and bakery products" group and the ”meat, poultry and fish" group, respectively, than the A. M. S. sample of families. The percentage of the total food expenditures which was spent on the "fruits and vegetables" group was about the same in all four samples of families. The percentage of expenditures on the "dairy products” group by the panel families was about 3. 5 percent, 1. 7 percent, and 4. 7 percent higher than it was for the B. L. S. , N. I. C. B. , and A. M. S. sample families, respectively. The percentage that panel families spent on the "other food at home" group was about 2. 4 percent lower than the B. L. S. sample of families and about the same as the N. I. C. B. and A. M. S. sample of families. The results indicate that the differences between the panel and the B. L. S. samples in the "dairy products" group and the “other foods at home" group were primarily due to an increase in consumption of the "dairy products" group and a decrease in consumption of the "other foods at home” group from 1952 to 1955-57. The A. M. S. sample families spent more of their food expenditures on the ”cereal and bakery products" group, and less on the "meat, poultry and fish" group and the ”dairy products” group, than the panel families. They spent about the same for the other groups of foods. This indicates that the farmers' spending patterns 183 for foods differ from those of the city workers. The low-income families customarily spent more of the food dollar for bread and cereal products than the high-income families. Smaller shares were spent by the farmers for dairy products because many farmers satisfy their own needs for some home-produced products. The above studies provide evidence that the panel data are sufficiently accurate and consistent for constructing a retail food price index. Although the accuracy and consistency of the data are of considerable importance in constructing index numbers, a sample cannot be expected to be representative unless an adequate number of items are included. A large sample of food items was selected in order to attempt to obtain reliable index numbers. The retail food price index of the M. S. U. Consumer Panel (the third index) was specifically designed to measure the effect of price changes on the cost of food in the 1955-57 "market basket" for Lansing urban families. The average size of the families in 1955-57 included in the index was estimated to be about 3. 2 persons, and their average family income was estimated at about $5, 452 after taxes. The content of the "market basket" was assumed to remain the same, so that the changes in cost from period to period are the result of changes in price alone. In the sampling procedure it was assumed that a relatively small number of properly selected items would be representative of an entire group. The estimate of general spending patterns of all 184 Lansing urban families was determined from the actual spending of a representative sample of the Lansing urban families. With this in mind, the data were first arranged in terms of the major groups of expenditures on foods by the panel families. The five major groups were constructed to represent the major com- ponents of family food expenditures. These five major groups (major components of the index) were further divided into 21 subgroups (sub- components of the index). Thus, a definite stratification was established in which each of the basic expenditure items was classified by type and assigned a proper subgroup. The main criteria used in selecting the items to be priced for each period were their importance and representativeness. The importance of the item in the total family food expenditures was used as a criterion, because the effect of price changes varied with the relative importance of the item. The representativeness of the item, in a group of items, was used as a second criterion, because the price change of the item selected for pricing in the index was used to represent the price changes of the unpriced items. Two factors were considered in determining expected similarities in price: fluctuations. One was the pattern of the past price fluctuation, and the other was the usage of the products. The basic expenditure items were grouped according to the above considerations, and one or more products were chosen to represent each subgroup. The expenditure for the selected items was then used to represent the combined expenditures of the group it represented. 185 The method of combining prices into the food price index was basically the technique defined by the Laspeyres' formula. The panel index numbers were weighted by base-year quantities. The simple mathematical notation for the formula was: I _ qupt t z qu0 100 In words this says: The index for time"t" equals the sum (2) of the products of the quantities (qo) of the items (fixed according to the panel family purchases at time "0") and the prices (pt) of the items for "t, " divided by the sum (2) of the products of the same quantities (qo) and the prices (p0) of the same items for the base period "0. 'q' Since the index is a percentage measure, the quotient of the two sums is multiplied by 100 to give an index number. This formula was also used for the major group and the subgroup index. In actual practice, however, a variation of this formula was used, namely, ZS: . poqo Z quO Thus, the weights used in the index were "value weights. " The summation of these value weights for the group of items makes it impossible to identify the quantity factors attached to each index item, despite the fact that the ”base price" and the "value weights" were both derived from the same quantity of products in the same period of time. 186 Quantity weights, therefore, were only implicit in the index structure. The panel food price index was based on the panel family expenditures for 1955-57. This expenditure represented the recent spending patterns of the Lansing urban families. Hence no considerations need necessarily to be taken of basic changes in items and weights. Therefore, the panel food price index did not require weight revision or linking. According to the main criteria used for selecting items, 131 food items were priced in the index. The items which were chosen were of outstanding importance in family purchases. These items in themselves represented the greater part of the food expenditures of Lansing families. The index was calculated using the average prices in each 4-week period. The 4-week average prices were compared with the prices obtained during the base period, 1955-57, average. The price relative of each item was calculated. Then, these price changes were put together to determine what happened to the prices on the average. To do this, each price relative was applied to its index weight to determine how much the price changes affected the total food expenditures. Seasonal variation in food consumption presents both practical and theoretical problems when index numbers are constructed. The panel food price index employed constant annual weights for fresh fruits and vegetables. Lack of prices for given items in given seasons was handled by estimation. The out-of—season prices of the items were either held constant until price again became available or assumed to have similar '1 to limitations not an exact subject to s ”sampling e: not the entir a fraction 0 Product sat: prices that keep the to‘ items prim their impoj Prices of f family too. seasonal p the aV’ErafL-l hr ge mlm Wive S Pr 0 reported . these Her the exper. 187 have similar price patterns as some year-round items. The M. S. U. Consumer Panel food price index is subject to limitations in measurement and use. This food price index is not an exact measurement. All the statistical measurements are subject to some kinds of limitations. One kind of limitation is ”sampling errors. " The index was based on a sample of families, not the entire population. In addition, the items selected are only a fraction of the total. The degree of error introduced into the index through product sampling depends mainly upon the amount of variations in prices that exists in the index within groups of items. In order to keep the total effect of these errors to a minimum, the number of items priced in the index was increased as the price variability and their importance in the all food index increased. For example, prices of fresh fruits and vegetables, which are important in the family food expenditure, change frequently and have different seasonal patterns for different items. Hence, in order to measure the average change in all fruits and vegetables satisfactorily, a large number of them were priced in the index. Another type of "error" occurs in the index because house- wives probably do not report the information exactly. Some may have reported lower prices than they actually paid, some higher, so that these ”errors of reports" may tend to cancel out. Nevertheless, the expenditure data and price data were compared and adjusted in 188 order to keep the effect of these ”errors" as small as possible. Another limitation is that of use of the index. The panel index is based on a specific sample. Use of the index to measure price changes for families other than the Lansing families will be appropriate only to the degree of similarity in spending patterns to the Lansing urban families included in the index. The index "market basket" is held constant so that price change alone will be reflected by the index. The index cannot be used to measure the level of consumption since it does not take into account the quantities of food consumed. Hence, to show changes in the “total cost of foods, " the "market basket" would have to be re-assessed at each period. The index was calculated for the all food at home group, for each major group, and for each subgroup. The indexes calculated were presented in Table 10 at the end of Chapter VI. The changes in prices of each food group, from one period to another, were measured by each group index in the table. . A graphic presentation of the price movements of each major group and subgroup of products was shown in Chapter VI. A brief analysis of each price movement followed each graphic presentation. . After the panel food price index was calculated, comparisons were made with other indexes. In the first place, compfarisons were made in order to evaluate the differences between the three different food price indexes, all for Michigan cities. The three different 189 indexes were the M. S. U. Consumer Panel food price index {based on the panel prices for Lansing), the B. L. S. food price index for Detroit (based on the store, sample prices for Detroit), and the N. I. C. B. index for Lansing (based on the store sample prices for Lansing). Comparison between the M. S. U. Consumer Panel retail food price index for Lansing and the B. L. S. retail food price index for Detroit indicated that these two indexes moved together fairly closely, with differences in price levels being due to different price quotations, different levels of base period prices, and different prices in the given periods. The correlation coefficient between the two indexes was . 82. However, the B. L. 5. index fluctuated with larger amplitude but less frequently than the panel food price index. Pre- sumably the greater frequency of variations in the panel index reflected more of the effects of the prices of "specialsH and, to some extent, changes in the quality of some of the items purchased by the panel families from week to week. A comparison of the food price index for Lansing between the panel series and N. I. C. B. series showed that the two indexes moved parallel to each other despite varying specifications of the food items. The correlation coefficient between the two indexes was . 81. As shown in Figure 32., there were wider fluctuations in the N. I. C. B. series than in the panel series. In addition to the differences between the store sample and panel sample prices, the wider fluctuation in the N. I. C. B. index probably resulted from the 190 combined effects of the greater relative importance of meat and fresh fruits and vegetables in the N. I. C. B. series than in the panel series. Secondly, comparison was made in order to test the effects of using the B. L. S. weights for Detroit and the M. S. U. Consumer Panel weights for Lansing (both based on the panel prices) in constructing a food price index for Lansing. The two indexes were calculated using the same base and same prices in the given period; but one was calculated using panel weights for Lansing (represented by the dotted line) and another was calculated using the B. L. S. weights for Detroit (represented by the dashed line). The two indexes, as shown in Figure 33, converged in the beginning of 1955. During this period the Lansing food price index would be the same whether the B. L. S. weights for Detroit or the panel weights for Lansing were used. From the 4th period to the 9th period of 1955, the dashed line moved above the dotted line. In this period the Lansing food price index would overestimate the cost of foods for Lansing families, if the Detroit B. L. S. weights were used in constructing the index. on the other hand, the B. L. S. food price index for Detroit would have biased upward the estimate of the cost of food for Detroit families, if the Detroit 1955-57 family spending patterns were the same as the panel 1955-57 family spending patterns. From the 9th period of 1955 through the 13th period of 1958, the index represented by the dotted line gradually moved above the index repre- sented by the dashed line. In this period the Lansing food price index 191 would underestimate the cost of foods for Lansing families, if it were based on the B. L. S. weights for Detroit. On the other hand, the B. L. S. food price index would have a downward bias in estimating the food cost for Detroit families, if their spending patterns for food in this period were the same as those of the panel families, It is very likely that the Detroit families, in the period of 1955-57, had spending patterns similar to those of the Lansing panel families in the same period. In this case, if the Detroit family spending patterns were the same as the Lansing families‘ in the period 1955-57, the B. L. S. index for Detroit would have underestimated the cost of foods for Detroit families from the latter part of 1955 to the end of 1958. Likewise, the B. L. S. index for Detroit would have overestimated, but very slightly, the cost of foods for Detroit. families from April to September of 1955. The M. S. U. Consumer Panel food price index has both . strengths and weaknesses. The obvious strength of the panel index is the method of obtaining information for prices and for weighting patterns. The panel method is more sensitive to change in individual behavior from time to time because the data are collected over time. The B. L. S. direct survey method cannot reflect the changes in con- sumption over time, since the information, which was obtained from interviews, covered only a very short period of time. The panel method can furnish more accurate information from weekly reporting diaries than the B. L. S. recall interviews. The panel index prov1des 192 a more accurate measure of the costs of food, since the spending patterns represent the actual purchases of the panel families and prices are those actually paid by the panel families. When comparing the weighting patterns actually used in this study, the panel index is less biased, since the weights used were the actual spending pattern of the panel families in 19 55-57. The B. L. S. food price index is biased more by its weighting pattern, since the index weights were based on the 1952 spending patterns. In selecting a preference for one index over the other, one would need to choose the type of error which is least harmful. Many researchers have suspected for some time that pricing on the basis of the most commonly purchased items would result, over the years, in a more realistic measure of price changes than pricing tied tightly to a specific quality of product (which represents only a small proportion of that actually bought. 193 BIBLIOGRAPHY Adelman, Irma. "A New Approach to the Construction of Index Numbers. " The Review of Economics and Statistics, Vol. XI, No. 3 (August, 1958). Agricultural Marketing Service, U. S. Department of Agriculture. Consumption and Utilization of gricultural Products. Vol. 5, pp. 65-68. Allen, R. G. D. Statistics for Economists. London: Hutchinson's University Library, pp. 100—116. Black, John D. and Mudgett, Bruce D. Research in Agricultural Index Numbers. New York: Social Science Research Council, March 1938. Bureau of Labor Statistics, U. S. Department of Labor. ”Interim Adjustment of Consumers' Price Index. " Monthly Labor Review. Vol. 72, No. 4 (April, 1951, pp. 421-430. Bureau of Labor Statistics, U. S. Department of Labor. ”Selection of Cities for Consumer Expenditures Survey, 1950. " Monthly Labor Review. Vol. 72, No. 4 (April, 1951), pp. 430-437. Bureau of Labor Statistics, U. S. Department of Labor. "Adjusted Consumers' Price Index: Relative Importance of Items. " Monthly Labor Review. Vol. 72, No. 6 (June, 1951). Bureau of Labor Statistics, U. S. Department of Labor. ”The Main Features of the. Revised Consumer Price Index. " Monthly Labor Review. Vol. 76, No. 2 (February, 1953). Bureau of Labor Statistics, U. S. Department of Labor. “Consumer Price Index. " Reprint of Chapter 9 from B. L. S. Bulletin No. 1168, Techniques of Preparing Major B. L. S. Sta- tistical Series. Bureau of Labor Statistics, U. S. Department of Labor. Food Price Specification Manual. Revised. February 2, 1953. Bureau of Labor Statistics, U. S. Department of Labor. ”Relative Importance of CPI Components. " Monthly Labor Review. July, 1959. Pp. 767-770. 194 Croxton, Frederick E. and Cowden, Dudley J. Applied General Statistics. 2nd ed. Englewood Cliffs, New Jersey: Prentice Hall, Inc. , 1956. Davis, J. S. "Standard and Content of Living. " American Economic Review. March, 1945. Kerger, Wirth F. "Distinctive Concepts of Prices and Purchasing Power Index Numbers. ” Journal of the American Statistical. Vol. XXXII (June, 1936), pp. 258-272. Fisher, Irving. The Making of Index Numbers. Boston: Houghton Mifflin Company, 1927. Frich, R. "Annual Survey of Economic Theory: The Problem of Index Numbers." Econometrica 4, 1936, pp. 1-38. Hicks, J. R. A Revision of Demand Theory. Oxford: The Clarendon Press, pp. 1-95 and 108-188. Hicks, J. R. Value and Capital. 2nd ed. London: Amen House, 1946, pp. 11-3‘8. King, Willford I. Index Numbers Elucidated. New York: Longmans, Green and Co. , 1930, pp. 164-65 and 185-86. Kozlik, Adolf. "Concepts of Plane, Standard, Level and Satisfaction of Consumption and Living. " Journal of Marketing, July, 1944, pp. 55-57. ‘ Leontief. ”Composite Commodities and the Problem of Index Numbers. " Econometrica. January, 1936. pp. 3?-59. Moss, Thomas N. “Some Relationships of Selected SociO—Economic Factors to Food Consumption and Expenditures, Lansing, Spring, 1950. " Unpublished Ph. D. thesis, Michigan State College, 1952. Mudgett, Bruce D. Index Numbers. New York: John Wiley and Sons, Inc., 1951, Chapter 4. v National Industrial Conference Board. Development of the New Market Basket. New York: National Industrial Conference Board, Inc., 1954, pp. 1-2. National Industrial Conference Board. The Market Basket for Un- surveyed Cities. New York: National Industrial Conference Board, Inc., 1955, pp. 1-3. 195 National Industrial Conference Board. Revised Consumer Price Index. New York: National Industrial Conference Board, Inc 1954, p. 304 ° 9 Quackenbush, Gerald G. “Demand Analysis from the M. S. C. Consumer Panel. ” Journal of Farm Economics, XXXVI, No. 3 (August, 1954). Rothwell, Doris P. ”Use of Varying Seasonal Weights in Price Index Construction. " Journal of the American Statistical Association. Vol. 53, pp. 66-77. Shaffer, J. D. “A Plan for Sampling a Changing Population over Time. “ Journal of Farm Economics, XXXVI, No. 1, 1954, pp. 153-163. Siegel, Irving H. “What Concepts Are Appropriate to Consumer Price Index?” Journal of Farm Economics, May, 1956, pp. 361-68. Snedecor, George W. Statistical Method. 4th ed. Ames, Iowa; Iowa State College Press, 1946. Stauber, B. R. ”The Parity Index and the Farm Expenditure Survey. “ Journal of Farm Economics, May, 1956, pp. 369-377. Stauber, B. R., Hole, R. F. and Peterson, B. S. "The January, 1959 Revision of the Price Indexes. " Agricultural Economic Research, Vol. XI, Nos. 2 and 3 (April-July, 1959),. pp. 33-80. ' Stigler, George J. The Theory of Prices. Revised ed. New York: The Macmillan Company, 1952, pp. 42-110. Teper, Lazor. ”B. L. S. Consumers’ Price Index and the A. M. S. Index of Prices Paid by Farmers for Family Living-e A Taxtaposition. Journal of Farm Economics, May ; 1956, pp. 3?8_399. Ulmer, M. J. The Economic Theory of Cost of Living Index Numbers. New York; Columbia Universnay Press. 1949. Wold, Herman and Jureen, Lars. Demand Analysis. New York: John Wiley and Sons, Inc. , 1953. Willard, Cochrane W. and Bell, Carolyn Shaw. The Economics of Epnsunfiation. New York: McGraw-~Hill Book Company, Inc., p. 72. 196 Woodbury, Robert Morse. ”Quantity Adjustment Factors in Cost-of Living Ratio. " Econometrica,'Vol. 8, No. 1 (January, 1940), pp. 322-332. 197 APPENDIX 198 APPENDIX 1 The M. S. U. Consumer Panel was not specifically designed for constructing a retail food price index. During the life of the panel food items were sometimes changed in order to get better information or more accurate reporting. Also, tabulations of prices on food items were changed on different occasions. Some data have been adjusted in order that every item in the index represented a uniform quality of food from the beginning to the end. In some cases several different. qualities of items which were combined in the panel data were separated into individual qualities, or in other cases several individual qualities were combined to represent a single item, for the index. The original panel data and tabulation methods did not insure the comparability of pricing items from one period of time to another. The following table shows the original data in the panel and the adjusted data for the index. Prices and Items in the Prices of Items Combined Panel Tabulation for the Index Beef steak 1 Beef steak 1951-54 Prices of all steaks 1951-58 Prices of the following individual items combined: Round and Swiss steak Sirloin steak Porterhouse and T -bone steak Other steaks Beef roast Beef roast 1951-54 Prices of all roasts 1951-58 Prices of the following ' individual items combined; Ham 1951-54 Prices of all hams Chicken 1951-53 Salmon 1951-53 1954-58 Tuna 1951-53 1954-58 Fish 1951-53 1954-58 Prices of broilers or fryers, ready to cook, fresh Prices of all salmon Prices of canned salmon Prices of all tuna Prices of canned tuna Prices of fish, other Prices of fish, other, cleaned and frozen Ham 199 Chuck roast, bone removed Chuck roast, bone in Rib roast, bone removed Rib roast, bone in Other roast, bone removed Other roast, bone in 1951-58 Prices of the following Chicken individual items combined; Ham, whole or half, bone removed Ham, whole or half, bone in Other hams, bone removed Other hams, bone in Ham, center slice Canned ham 1951-58 Prices of the broilers and Salmon fryers, ready to cook, fresh 1951-58 Prices of canned salmon Tuna 1951-58 Prices of canned tuna Fish 1951-58 Prices of fish, Other , cleaned and frozen 200 APPENDIX 2 Retail Prices of Individual Foods, Average of the Three Years 1955-57, M. S. U. Consumer Panel 1955-57 1955-57 Food and unit average Food and unit average in cents in cents Cereal 8: bakery products Other meats , Wieners & frankfurters lbs. 50. '7 ————C€real Bolo na lbs 51 9 Wheaties lbs. 32.4 g ° ' Other cold cuts lbs. 72.. 8 Corn flakes lbs. 30' 0 Prem S am Treet lbs 52 8 Corn sugar frosted flakes lbs. 44. 4 ' p ' ° ° Rolled oats lbs. 14. 9 Veal cutlet lbs. 70. 1 White flour lbs. 9. 4 L b lb . 65. 6 Corn meal lbs. 11. 6 am 5 White cake mix lbs. 28. 6 Poultry Rice lbs. 26. 6 Fryers lbs. 43. 9 Biscuit mix lbs. 17. 7 Roasters lbs. 49. 5 Spaghetti or macaroni lbs. 23. 2 Turkey, ready-to-cook ' f lb . 60. 7 Bakeerroducts rozen S Bread, white lbs. 18. 1 Fish. sea fOOdS Diet bread lbs. 2.5. 1 Tuna lbs. 73. 3 Buns, hamburg . 08 ea. 35. O . . Fish sticks lbs. 64. 5 Soda crackers lbs. 28. 6 . , , .3... Other fish lbs. 45. 8 Cookies --pla1n 8r sugared .05 ea. 41. 4 O ysters lbs. 107. 3 Doughnuts--frosted and Shrim lbs 100 6 sugared .07 ea. 46. 8** p ° ° Cherry pies lbs. 37. 1 Dairy products come cake lbs’ 56' 0 Milk, fresh (Vit.D) qt. 20.6 Meat, poultry and fish Milk, fresh (Reg. Past. ) qt. 20. 7 M t Milk, evaporated lbs.- 15. 7 ea Milk, powdered, skim lbs. 40.1 Beef . Ha b lb 39 6 Coffee cream .5pt. Z3. 7 m urger s. ' Cheese (Natural American) lbs. 61. 6 All roast lbs. 58. 4 All t k lb 93 7 Cheese (processed, 8 ea 3. ' American) lbs. 50. 9 Pork Cottage cheese lbs. 28. 2 Pork chops lbs. 71. Z Butter lbs. 67. 2 Pork roast lbs. 48. 1 Ice cream, pre-packaged pt. 22. 4 IBI:k The estimates of the out-of—season prices are included in the average. >3 >l< Cookies--if weights were not given, . 05 was used for each. Doughnuts--if weights were not given, . 07 was used for each. Buns--hamburg--if weights were not given, . 08 was used for each. Price in table refers to price per pound. 203 APPENDIX 3 The M. S. U. Consumer Panel retail food price index employed constant annual weights for fresh fruits and vegetables. Lack ‘of prices for given items in given seasons was handled by estimation-- either by holding constant (using annual averages of the prices in season for the prices of out-of-season periods) until prices again be- come available, or by assuming the same price changes for out-of- season items as for year-round items. ‘ The out-of-season prices of the items which were used in the index were estimated as follows: The out-of-season prices of grapes and peaches were estimated by using the apple prices. The out-of—season prices of grapefruit were estimated by using the prices of oranges. Grapes and peaches have price movement patterns similar to apples. Grapefruit have price movement patterns similar to oranges. Apples and oranges are year-round items. The same reasons were applied in using tomato prices to estimate the price of green beans. However, no other items, according to the patterns of price movements, can be used properly for estimating the out-of-season prices of strawberries, watermelons and "cantaloupe and muskmelon. " Therefore, the annual averages of the prices in season were. used for the prices of the out-of-season prices for water- melons and "cantaloupe and muskmelon. " Since both strawberries and lemons used to have'high prices in May and low prices in June, the lemon prices were used to estimate the out-of-season prices of strawberries. 204 APPENDIX 4 1951 1952 1953 1954 Date Week Date Week Date Week Date ' Week lZ-3lto 1-5 1 12-30to 1-5 1 12-23to 1-3 1 12-27to 1-2 1 1-7 1-13 2 1-5 1-12 2 1-4 1-10 2 1-3 1-9 2 1-14 1-20 3 1-13 1-19 3 1-11 1-17 3 1-10 1-15 3 1—21 1-27 4 1-20 1-25 4 1-18 1-24 4 1-17 1-23 4 1-28 2-3 5 1-27 2-2 5 1—25 1-31 5 1-24 1-30 5 2—4 2-10 5 2-3 2-9 5 2-1 2-7 5 1-31 2-5 5 2-11 2-17 7 2-10 2-15 7 2-8 2-14 7 2-7 2-13 7 2-18 2-24 8 2-17 2-23 8 2-15 2-21 8 2-14 2-20 8 2-25 3—3 9 2—24 3-1 9 2-22 2-28 9 2-21 2-27 .9 3—4 3-10 10 3-2 3-8 10 3-1 3-7 10 2-28 3—5 10 3-11 3-17 11 3-9 3-15 11 3-8 3-14 11 3-7 3-13 .11 3-18 3-24 12 3—15 3-22 12 3-15 3-21 12 3-14 3-20 12 3-25 3—31 13 3—23 3—29 13 3_22 3-28 13 3-21 2-27 13 4-1 4—7 14 3-30 4-5 14 3-29 4—4 14 3—28 4—3 14 4-8 4-14 15 4-5 4-12 15 4-5 4—11 15 4-4 4-10 15 4-15 4—21 15 4-13 4-19 15 4-12 4-18 15 4-11 4-17 15 4-22 4-28 17 4—20 4-25 17 4-19 4-25 17 4-18 4-24 17 4-29 5-5 18 4-27 5-3 18 4-25 5-2 18 4-25 5-1 18 5—5 5-12 19 5-4 5-10 19 5-3 5-9 19 5—2 5-8 19 5-13 5-19 20 ‘ 5-11 5-17 20 5-10 5-15 20 5-9 5—15 20 5-20 5-25 21 5-18 5-24 21 5-17 5—23 21 5-15 5—22 21 5-27 5-2 22 5-25 5-31 22 5-24 5-30 22 5-23 5-29 22 5—3 5-9 23 5-1 5-7 23 5-31 5-5 23 5-30 5-5 23 5-10 5-15 24 6-8 5-13 24 5-7 5-13 24 5-5 5-12 24 5-17 5-23 25 5-15 5-21 25 5-14 5-20 25 5-13 5-19 25 5-24 5-30 25 5-22 6-28 25 5-21 5-27 25 5—20 5-25 25 7—1 7-7 27 5-29 7-5 27 6-28 7-4 27 5-27 7-3 27 7-8 7-14 28 7—5 7-12 28 7-5 7-11 28 7-4 7-10 28 7_15 7-21 29 7—13 7-19 29 7—12 7-18 29 7-11 7—17 29 7—22 7-28 30 7-20 7-25 30 7-19 7-25 30 7-18 7-24 30 7-29 8-4 31 7-27 8-2 31 7-25 8-1 31 7-25 7-31 31 8.5 8-11 32 8-3 8-9 32 8-2 8-8 32 8-1 8-7 32 8-12 8—18 33 8-10 8-16 33 8—9 8-15 33 8-8 8-14 33 8-19 8-25 34 8-17 8-23 34 8-16 8-22 34 8-15 8-21 34 8-26 9-1 35 8-24 8-30 35 8-23 8-29 35 8—22 8-28 35 9-2 9-8 35 8-31 9-5 35 8—30 9-5‘ 35 8-29 9-4 35 9-9 9-15 37 9-7 9-13 37 9-5 , 9-12 37 9-5 9-11 37 9-16 9-22 38 9-14 9-20 38 9-13 9-19 38 9-12 9-18 38 9-23 9-29 39 9—21 9-27 39 9-20 9-25 39 9-19 9-25 39 9-30 10-5 40 9-28 10-4 40 9—27 10—3 40 9-25 10-2 40 APPENDIX 4. - - Continued 205 1951 1952 1953 1954 Date Week Date Week Date Week Date Week lO-7t0 10-13 41 10-5 UDIO-ll 41 10-4 “310-10 41 10-3 H>10~9 41 10—14 10-20 42 10-12 10-18 42 10-11 10-17 42 lO—ll 10-16 42 10-21 10-27 43 10-19 10—25 43 10—18 10-24 43 10-17 10-23 43 10-28 11- 44 10-26 11-1 44 10-25 10-31 44 10—24 10430 44 ll~4 11-10 45 11-2 11-8 45 11-1 11-7 45 10-31 11-6 45 11-11 11-17 46 11-9 11—15 46 11-8 11—14 46 11-7 ll-l3 46 11-18 11-24 47 11-16 ll~22 47 11-15 11-21 47 11-14 11-20 47 11-25 12-1 48 11-23 11-29 48 11-22 11-28 48 11-21 11-27 48 12-2 12~8 49 11—30 12-6 49 11-29 12-5 49 ll-28 12-4 49 12-9 12-15 50 12-7 12-13 50 12—6 12—12 50 12-5 12-11 50 12-16 12-22 51 12-14 12-20 51 12-13 12-19 51 12-12 12-18 51 12-23 12-29 52 12-21 12-27 52 12—20 12~26 52 12-19 12-25 52 12-26 1-1 53 1955 1956 1957 1958 Date Week Date Week Date Week Date Week 1-2to -8 l -lto -7 1 12-30to l~5 l 12—29to 1-4 1 1-9 1—15 2 -8 1—14 2 1—6 1-12 2 1-5 1—11 2 1-16 1—22 3 1—15 1-21 3 1-13 l-l9 3 1-12 1-18 3 1-23 l~29 4 -22 1—28 4 1-20 1-26 4 1-19 1—25 4 1-30 2-5 5 1-29 2-4 5 1—27 2- 2 5 1-26 2-1 5 2-6 2~12 6 -5 2-11 6 2—3 2- 9 6 2—2 2-8 6 2—13 2-19 7 2-12 2—18 7 2—10 2-16 7 2-9 2—15 7 2—20 2-26 8 2-19 2~25 8 2-17 2— 23 8 2-16 2~22 8 2~27 3-5 9 2s26 3-3 9 2-24 3- 2 9 2-23 3~l 9 3-6 3-12 10 3-4 3-10 10 3-3 3— 9 10 3-2 3~8 10 3-13 3—19 11 3~ll 3~17 11 3-10 3- 16 11 3—9 3-15 11 3-20 3—26 12 3-18 3-24 12 3-17 3- 23 12 3-16 3sZ? 12 3-27 4-2 13 3-25 3—31 13 3-24 3 3O 13 3-23 3-29 13 4-3 4~9 14 4-1 4~7 14 3-31 4- 6 14 3—30 4-5 14 4-10 4-16 15 4-8 4-14 15 4-7 4 13 15 4-6 4-12 15 4-17 4—23 16 4-15 4:21 16 4~14 4e20 16 4-13 4-19 16 4~24 4-30 17 4-22 4-28 17 4-21 4-27 17 4»20 4426 17 5—1 5-7 18 4-29 5~5 18 4 28 LM4 18 4~27 5.3 18 5-8 5-14 19 5-6 5~12 19 5-5 5-11 19 5-4 5~lU 19 5-15 5—21 20 5—13 5-19 20 5—12 5-18 20 5-11 5 17 20 APPENDIX 4. --Continued 206 1955 , 1956 1957 1958 Date Week Date Week Date Week Date Week 5-22 to 5-28 21 5-20 to 5-26 21 5-19 to 5-25 21 5-18 to 5-24 21 5-29 6-4 22 5-27 6-2 ‘ 22 5-26 6-1 22 5-25 5-31 22 6-5 6-11 23 6-3 6-9 23 6-2 6-8 23 6-1 6-7 23 6-12 6-18 24 6-10 6-16 24 6-9 6-15 24 6-8 6-14 24 6-19 6-25 25 6-17 6-23 25 6-16 6-22 25 6-15 6-21 25 6-26 7-2 26 6-24 6-30 26 6-23 6-29 26 6-22 6-28 26 7-3 7-9 27 7-1 7-7 27 6-30 7-6 27 6-29 7-5 27 7-10 7-16 28 7-8 7-14 28 7-7 7-13 28 7-6 7-12 28 7-17 7-23 29 7-15 7-21 29 7-14 7-20 29 7-13 7-19 29 7-24 7-30 30 7-22 7-28 30 7-21 7-27 30 7-20 7-26 30 7-31 8-6 31 7-29 8-4 31 7-28 8-3 31 7-27 8-2 31 8-7 8-13 32 8-5 8-11 32 8-4 8-10 32 8-3 8-9 32 8-14 8-20 33 8-12 8-18 33 8-11 8-17 33 8-10 8-16 33 8-21 8-27 34 8-19 8-25 34 8-18 8-24 34 8-17 8-23 34 8-28 9-3 35 8-26 9-1 35 8-25 8-31 35 8-24 8-30 35 9-4 9-10 36 9-2 9-8 36 9-1 9-7 36 8-31 9-6 36 9-11 9-17 37 9-9 9-15 37 9-8 9-14 37 9-7 9-13 37 9-18 9-24 38 9-16 9-22 38 9-15 9-21 38 9-14 9-20 38 9-25 10-1 39 9-23 9-29 39 9-22 9-28 39 9-21 9-27 39 10-2 10-8 40 9-30 10-6 40 9-29 10-5 40 9-28 10-4 40 10-9 10-15 41 10-7 10-13 41 1056 10-12 41 10-5 10-11 41 10-16 10-22 42 10-14 10-20 42 10-13 10-19 42 10-12 10-18 42 10-23 10-29 43 10-21 10-27 43 10-20 10-26 43 10-19 10-25 43 10-30 11-5 44 10-28 11-3 44 10-27 11-2 44 10-26 11-1 44 11-6 11-12 45 11-4 11-10 45 11-3 11-9 45 11-2 11-8 45 11-13 11-19 46 11-11 11-17 46 11-10 11-16 46 11-9 11-15 46 11-20 11-26 47 11-18 11-24 47 ll-l7 11-23 47 11-16 11-22 47 11-27 12-3 48 11-25 12-1 48 11-24 11-30 48 11-23 11-29' 48 12-4 12-10 49 12-2 12-8 49 12-1 12-7 49 11-30 12-6 49 12-11 12-17 50 12-9 12-15 50 12-8 12-14 50 12-7 12-13 3) 12-18 12-24 51 12-16 12-22 51 12-15 12-21 51 12-14 12-20 51 12-31 52 12-23 12-29 52 12-22 12-28 52 12-21 12-27 52 12—25 207 APPENDIX 5 fllrvfirc‘f/s 5 2 / .353}. f, o—J-"A—A AA (Again FAICHIGAN STATE UNIVERSITY WEEKLY CONSUMER roon PURCHASE DIARY 'l’his diary is :8“ r‘gqprigggg all food purchases for the week of Sunday ............................ through Saturday ....... 0 Pi}: ..... 1.8.3.8. 1. May we emphasize that each of your diaries is important to us, whether your food purchases are many or few. Your diaries will be of most value if made out accurately and returned promptly — every week. 2. We suggest that you enter food items in the diary each day as you make the purchase. 3. If a food item that you use is home-grown or a gift, show this by writing ”home— grown" or ”gift" in the price column. 4. If you don't know under which heading to enter a food item, you can list if in one of the blank spaces on page 19. 5. At the end of the week check through the diary to make sure you haven't forgotten any purchase or made any incomplete entries. 6. As you are checking the diary also V the squares IE] None) if appropriate. 7. If you want any information, call us at — number ED. 2-1511, extension 3030. INDEX no: use nos BAKED GOODS . . . . 13-15 FATS and OILS ........ 3 SOUPS .............. I 1 BEVERAGES .......... 17 FISH and SEA FOOD. . . . IO SUGAR, SWEETS, NUTS .16 BABY FOODS ......... 12 FRUITS ........... 4 & 5 VEGETABLES ....... 6 8: 7 CANDY ............. 16 GRAIN PRODUCTS. . . 13-15 VITAMINS, MINERALS . .17 COOKING AIDS ....... 17 JAM and JELLY ........ II VITAL DATA DAIRY PRODUCTS. . . 2 8s 3 MEAT ............. 8 & 9 Questions .......... I9 EGGS ............... 10 POULTRY ............ 10 DIRECTIONS .......... 18 WHAT YOU CAN EARN BY KEEPING A DIARY - If you return the diary for 52 weeks or more without missing a week, you earn 40 points for each diary returned in the sequence. or If you return the diary for 12 to 51 weeks without missing a week, you earn 35 points for each diary returned in the sequence. or If you return the diary for 5 to 11 weeks without missing a week, you earn 25 points for each diary returned in the sequence. or If you return the diary less than five weeks in a row, you earn 1 0 points for each diary. PLUS 1. A bonus of 5 points for each diary returned on time (postmarked before Tuesday noon of the following week). 2. A bonus of 70 points if you return every diary on time for a year. 3. A bonus of 10 points for each diary returned during July and August. 4. A bonus of 5 points for each diary returned after returning 52 diaries. You can earn 2500 points the first year and 2760 points in each additional year. I2) DAIRY PRODUCTS MILK NONE '3 .— fl-ll SB Numlor of Prict par Total Whore ICE FRESH 1 100 on. 99 Gal. 09. vs Gal. Amt. Paid Brand Purchaud' Ho. Homogenized—Vit. D. 1 I 10 7 __ 3 Pre , _ j ‘ on Multiple Vitamin Milk — (Brown Bottle) 1 I I 1 2 She ~~ *— —— o Homogenlzed—Plaln 1 120 v — _ ; Dai Regular Pasteurized I 130 ‘" CHE Jersey or Guernsey 1 I40 ,, —_ Nat Buttermilk I 150 P re Chocolate 1160 S . __.__. __. w1 Sklm Milk 1170 *0“ Sour Milk, Yoghurt, etc. 1 180 Cre Egg Nog, etc. 1 181 Otl’ Other Milk 1 190 \ Half 8. Half 1192 Cot N mb Pl Tot I Where § CREAM NONE D I400 of :2 ”In?" per ‘56"?an Amt. :ald Brand Purehaud' Coffee Cream 1410 3 § 3 Whipping Cream—bottle or carton 1421 ‘3 F I Whip. Cream—can (Reddi-Whip, etc.) 1422 E fl Sour Cream— '/2 Pint Size 1430 fl Sour Cream—12 Oz. Size 1430 & l Loft 8 v ’For Fresh Milk and Cream — Please indicate from whom it was bought in the last column as follows: \w: 1. If delivered by milkman 4. If bought from coin-operated milk dispenser & 2. If bought from grocery store 5. If bought from other source 0» 3. If bought from cash and carry specialized dairy store \ \ OIL! N mb w.“ m t T l I W!- CANNED (Liquid) 1200 .i’cmi.’ sci: Ca: Am: :ald Brand hunts?“ Tm Evaporated~Unsweetened I 210 0 Ta 0 . Condensed—Sweetened 1 220 i? T a. GI: Canned Skim 1230 T: R s °q Canned—Baby Formulas 1240 g Sch 0 Number Prlco Total Whore ': m DRIED I 300 of Pounds par Pound Amt. Pald Brand Purchased .3 Powdered—Skim Milk 1310 & Powdered—Whole Milk 1320 & Powdered—Baby Formulas I330 & Ice Cream Mix 1340 \ Sherbet Mix 1341 “W on Malted Milk Powder 1321 :; Pream (Powdered Cream) I350 51‘ s: #z, a;i____ -i,__. _ ,, __. ._ __ __.‘ -..:".¢’ 1 . ~ 2 I. "' 1;. ..: .. l. .. O .-.\ V ——_*__ _ O \ \ \ \ \ \ \ \\\\ DAIRY PRODUCTS (cont.) (3i ms: :3 ICE CREAM NONE D 1500 will??? OG‘OI. piiic°12°kol £331.... Brand p.33... Hand Packed Ice Cream 1510 Pre-Packaged Ice Cream 1520 Other Ice Cream 1530 Sherbets and Ices 1540 Dairy Queen, Frostie, etc. 1550 CHEESE NONE D T3322? pol-P23“ Amorxt‘tfll’ald Pufizstd Natural American (Cheddar, etc.) 1610 Processed American (Velveta, etc.) 1620 Swiss Cheese 1730 Cheese Spread 1720 Cream Cheese (Philadelphia, etc.) 1750 Other Cheese I740 Cottage Cheese 1820 FATS AND OILS FATS NONE D 2100 .2323. p.323... 111:3:le will?“ Butter 21 10 Oleomargarine 2120 Lard 2130 Swiftning 2131 Vegetable Shortening (Crisco, Spry, etc.) 2140 _Other Fats (name kind) 2150 OILS NONE [:1 2200 9:13:32: p315?» 11.133le will?“ Cooking Oils 2210 Mayonnaise 2220 Salad Dressing 2221 Roquefort Dressing 2222 Salad Oils, etc. 2230 Other Oils 2240 _French Dressing, etc. 2231 Eandwich Spreads, Tartar Sauce 2242 _Whips (name kind) 2250 For all items on pages 2 and 3, except fresh milk and cream, indicate where purchased as follows: SR -— Shop-Rite Store W— Wrigleys (Big Bear) Store G—Any other grocery store or meat market not listed A — A&P Store B -— Barley or Junedale Store K -— Kroger Store M —- Market Basket (National) Store S —- Schmidts Store O—Any place other than grocery store or meat market, i.e., drug store, restaurant, tarm, dairy store, etc. I" FRUITS m o... AMOUNT aouom th ' ' ' ‘ ‘— EXAMPLES I s '12:. "‘32:?" ...... t . E z BERRIES NONE E) 3100 3 medium 2-8 oz. iars Amt.Pald E 3 I: 5 5 r . O Blueberrles qts. fresh 31 10 —— Cranberries 312C? . '1 Currants qts. fresh 31 21 ——I Dewberries and Blackberries qts. fresh 3130 —~: Raspberries qts. fresh 31g ..1 Strawberries qn. fresh 3150 ’ —— Other Berries (name kind) 3T60— —! Berry Juice (kind)? W—— — —3I70— _ —: CITRUS NONE 13 3200 '1 Grapefruit ng—Igh . ~372107 _‘, Grapefruit Juice 3279 _, Lemons no. fresh 3220 — Lemonade 3225— ——1 Lemon Juice 3229— '— a,_._ ,, *- _ - z, —t Limes no. fresh 3230 ., Lime Juice 323? —l Jim“ 3,233 ‘. Oranges no. fresh 3240‘ _| Frozen Orange Juice 3243‘ _l Orange Juice (not frozen) 324: ‘1 Orange Drink (Hi-C, etc.) 3246 —l 7 Tangerines no. fresh 3250 _ Tangerine Juice 3251—2 ‘ Other Citrus (name kind) 3260 ‘ Other Citrus Juice (name kind) 3270 ‘ Mixed Citrus Fruit 3290— . ‘1 Mixed Citrus Juices 3299 ‘1 OTHER FRUITS NONE 1:) 33-35 ‘ Apples 31W ‘ Applesauce and Applebutter 351‘ ‘| Apple Cider 3.3—1? 7 Apple Juice 3319 \t Apricots 3320 7 1 ln reporting Fruits and Vegetables please indicate the actual amount bought in weight or liquid measure, EXCEPT for those fresh Fruits and Vegetables indicated with qts., no ‘I heads, bunches after the name of the product it is all right to express the amount bought in these units. Ple FRUITS (cont.) ‘5’ “ECK ON! AMOUNT eouom _I'°"‘ _' ' ' EXAMPLES . . 3 plnts 3-6 as. cans :5. g E 1.12 oz. pkg. 2 lb. s 3 at. mu, 3 5 3 5 OTHER FRUITS CONT. 311W (0st.) home 9'°W" Amt-Paid : 3 ‘z u Apricot Nectar 3329 2* : — ‘ Avocados 3330 __ Bananas Ibs.lresh 3340 —. —Cherries—Maraschino 33750 i : — Cherries—Sour 3351 .— Cherries—Sweet 3352 h— — Dates 3360 — : Figs 3370 _ Grapes 3380 *_7 Grape Juice (single strength) 3389 ‘ Grape Juice (concentrated) 3388 ‘ : Cantalope and Muskmelon no. fresh 341 I _ — Watermelon 3412 __ Nectarines 3420 h “v Olives 3435 § Persimmons 3430 m — Peaches 3440 Pears 3450 Pineapple 3460 Pineapple Juice 3469 Plums 3470 ‘ Prunes 3480 ~ Prune Juice 3489 Raisins 3510 Rhubarb 3520 Hawaiian Punch Base 3540— Mixed Fruits (except citrus) 35907 — 1 Fruit Cocxtail 3591 Fruit Pie Mix (kind) 355' ‘l Mixed Fruit Juices 3542 Fruit Gelatin Salad—prepared 3570 Powdered Juice (kind) 3550 —— —| Candied Fruit (kind) 3560 —| Fruit Pickles (kind) 3580 I _ All Other Fruit Ju1ce (kind) 3549 I 1 _l i All Other Fruit (name kind) 3530 I I 1 Please don't forget to enter home grown, and gift items. ‘6’ VEGETABLES l CHECK ONE AMOUNT BOUGHT Fresh ' ‘ ' ’ EXAMPLES 1. t ‘ GREEN LEAH" '-.'-:';:.°:-.':::°~ ‘."'.:.‘.::.'" . z VEGETABLES NONE 1:1 41'43 V2 dos: ears I bunch Amt. Paid 5 2 '5.- s3 5 Brussel Sprouts 41 10 — Cabbage lbs.fresh 4120 _ Cabbage Salad 4122 ‘— Sauerkraut 4121 _ Celery bunches fresh 4130 _ Celery Cabbage bunches fresh 4140 _ Endive, Chicory, Escarole 4160 _ Greens—Beet, Mustard, etc. 4240 _ Lettuce—Head heads rml. 4210 '— Lettuce—Leaf 4220 _ —felruce —— Bibb 4221 — Parsley, Swiss Chard, Water Cress 4250 _ Spinach 4260 _ Mixed Leafy Vegetables 4290 — Other Leafy Vegetables 4300 _ Fresh _ 15 c 1: GREEN AND YELLOW W E a 5 3 VEGETABLES NONE [:1 44-45 mow aoucm 4...”... E .3 -= 3 6 Artichokes 4410 _ Asparagus 4420 _ Beans — Lima 1 D check if in pod 4430-1 Beans—Snap 4440 _ Bean Sprouts 4450 _ Broccoli 4460 Carrots bunches fresh 4470 Corn—Sweet no. of ears fresh 4480 Peas 1 Cl check if in pod 4530-1 Peppers no. fresh 4540 Pumpkin tbs. fresh 4550 Squash |bs.fresh 4560 Soya Steak and Chaplets 4951 Mixed Green and Yellow Vegetables 4590 Other Check prepackaged for fresh Fruits and Vegetables if the product has been wrapped or placed in a consumer size package with a specific price indicated for each package before the customer enters the store. \_\ 4L \ ‘ ‘77" \ ,\. VEGETABLES (cont.) ‘7’ CHE“ ONE AMOUNT BOUGHT Fresh ' ' ' ‘ ALL OTHER 4.473.733?“ 3 g g VEGETABLES NONE D ‘7'” "It..." Tanner»: '33" 4.3.3:... E 3 5 3 g Beans—Navy, Baked, White 4701 ——_ Pork and Beans 4701 ‘ ~ Beans—Kidney 4703 hI~ — — _ Beets 4710 _____ Cauliflower heads in... 4720 _' — —_'— Cucumbers no. trash 4731 I- 7— _ — — Cucumber Pickles 4733 — ~ — —_ Relish 4734 — Egg Plant he. end. 4740 __ 7’ Garlic 4750 § _ — ‘ — Horseradish 4760 - : Mushrooms 4780 — _ — - _ Onions—Mature 4811 — _ Onions—Green bunches fresh 4812 Parsnips 4830 — —. ‘ Pimentoes 4770 - : _ _ — I Michigan Potatoes 4841 ~ — _ — — — Maine Potatoes 4842 — ~ _~ — Idaho Potatoes 4843 — _ - _ _ California Potatoes 4844 _ — _ — _ Potatoes grown in Other States 4845 — Potatoes (Don't know state) 4846 — Potatoes—French Fries 4866 _7 Potato Chips 4867 — _ Potato Sticks 4868 — _ — _ Potato salad 4869 — : Mashed Potatoes or Patties 4865 i Sweet Potatoes and Yams 4570 _— — Radishes bundles fresh 4850 — _ Tomatoes 4871 _ I x \‘\ \ \ \ \_ \“‘\‘,\ \ \ Tomato Catsup, Paste, Sauces 4872 I _ Tomato Juice 4873 — Turnips and Rutabagas 4880 I _ — Prepared Veg. Gelatin Salad 4983 I —- I Mixed Vegetables 4950 I I _ I I Chop Suey, Chow Mein, without meat 4974 I I I I I__ _ Mixed Vegetable Juice 4960 I I I I I I ‘ l—I-l—— Other Vegetables (name kind) 4900 I l I Be sure to fill in both the "Amount Bought" and “Total Amount Paid" check the method of preservation. columns, as well as (3) MEAT 51 Number of Price Total Frozen Where BEEF NONE D 5100 Lbs" 02:. p0! Pound Amt. Paid When Purchased Canned Beef Corned Beef Chipped Beef Ground Beef, Hamburger V Ground Round Steak, Lean Gr. Beef Beef Liver and Baby Beef Liver Heart, Tongue, other Organ Parts Chuck Roast (Pot Roast) Rib Roast $9,232: fn°m°v°d Other Roast g B :32: It‘emmed Round and Swiss Steak Sirloin Steak Porterhouse and T-Bone Steak Other Steak Stewing Beet (Boneless) Boiling Beef or Short Ribs All Other Beef (name kind) Check if, Number of Price Total ' Frozen Whore PORK NONE D 5300 Lbs., 018. per Pound Amt. Paid £222, Purchased Bacon 531 1 Canadian Bacon 5312 Canned Pork 5320 Chops 5330 Steaks 5335 Ham—Center Slice 5340 Ham—Whole or Half ; B 33:: fif‘mved 534 Ham—Canned 5343 Ham—Other g ! 22:: fiemved 534 Picnic Ham, Cured Butts 5346 Pork Liver 535i _- Heart, Tongue, other Organ parts 5352 Roast—Fresh 5370— Sausage—Link 5380 Sausage 5381 Spareribs 5382 Side or Salt Pork 5391 '— Other Pork (name kind) 5390 ——r_- MEAT (cont.) ‘9’ I I 51 $3 . Checklt Number of Price Total Frozen Where lAMB—MUTTON NONE D 5200 “33., 01:. per Pound I Amt. Paid $222, Purchased Chops-Steaks 5220 I Roast (Leg, etc.) 5260 _I Other Lamb-Mutton (kind) 5280 VEAI. NONE E] 5400 T335523 perPPlgioind AnirP'ald crhlijil" Pu‘rvdhz:ed Cutlets, Chops, Steaks 5420 Ground Veal 5430 Call Liver 544] City Chicken 5443 Roast (1) [El] 22:: Emma 546 Stewing, Soup Veal 5470 Other Veal (name kind) 5490 OTHER MEAT AND “.2222: MEAT MIXTURES NONE g 5500 22:12:: ..:::... .2222. ..:::. Wieners and Franks, etc. 5510 Bologna—Ring or Large Round 551 1 Other Cold Cuts 5513— Other Cold Cuts 5513 Prem, Spam, Treet, etc. 5514 Rabbit, Domestic 5520 Venison and Other Game Animals 5530 Chop Suey Meat and Kabobs 5540 Bouillon Cubes _’ 5596 Beef Stew 5594 Chile Con Carrie 5591 Hash 5592 Mincemeat 5593— Meat Balls and Spaghetti 5595 Ravioli and Tamales 55958- Chop Suey, Chow Mein with Meat 5599 Potted Meat 5560 Meat Spreads 5570" —Pork and Beans 4701 -3 :Other 5550 For all items on pages 8 and 9 indicate where purchased as follows: 58 -- Shop-Rite Store W—Wrigleys (Big Bear) Store G—Any other grocery store or meat market not listed 0 - Any place other than grocery store or meat market, i.e., delicatessen, farm, city market, tood locker plant, etc. A — A&P Store I —Bazley or Junedale Store K —- Kroger Store M -—Market Basket (National) Store S —— Schmidts Store —e “0’ POULTRY, FISH, EGGS Check One II Check One I 8 8 C S I l I 3 C N mb Prlc Tot I 1: 3 .x 1.3 c POULTRY NONE 1:] 5600 ”... .. ,,.,° ...:.. g g 5 .33 g 1.; e1 5 a ‘5’ CHICKEN Lbs., 018. Pound Pald i 5:: d. 2 g EEI E E a Broilers or Fryers 5611 I Roasters 5612 Stewing 5613 Barbecued Chicken 5614 _ TURKEY 5620 DUCK 5630 __ OTHER POULTRY (kind) 5640 __ __ Game Birds 5650 MIXTURES—CHIEFLY CHICKEN 5690 Chicken Noodle Dinner 5691 Chicken a la King 5692 Chicken Chop Suey, etc. 5694 ___ Other (kind) 5696 __ __ Numb r Prl Total ’1 ” Check it EGGS NONE D 5700 of Doz:n per 0210" Amount Paid Size Grade Ungraded Check One Check One I I 8 C l =1 ’ 8‘ 4 Number at Price Total c 3 1 :0; .3 2! '2 :13 Po nd Amoun 0 1 T m 3 c .2 3 FISH AND SEA rooo .../2a“... .232. ....’ '5 a .2 8:5 .. 3 v; 5 NONED u. u. u a a... 3300 0 Tuna 581 1 Salmon 5812 Fish Sticks 5814 Other Fish (kind) 5813 __ Lobster, Lobstertail 5819 Oysters 5820 __ Oyster Stew 5821 __ Scallops 5830 __ Shrimp 5840 Tuna Pie or Casserole 5891 __ Sardines in Oil 5892 __ __ Sardines in Sauce 5893 __ .._. nifi.al.~‘_1- JAM, JELLY, soup, MEAT PIES, DINNERS “H JAM. JELLY. PRES ERVES, ETC. None :1 32:22:22 “22:25? ..::‘zi... ¥Blueberry 31 10-5 hCranberry 3120-5 Currant 3121-5 Raspberry (red or black) 3140-5 Strawberry 3150-5 Other Berry (kind) 3160-5 Orange 3240-5 Apple Jelly 3310-5 Apricot 3320-5 Cherry 3351-5 Grape 3380-5 Peach 3440-5 Pineapple 3460-5 Plum 3470-5 Mixed Fruit 3590-5 Other Fruit (kind) 3530-5 soup, mom, e.,... ”a... I,” CONSOMMEI ETC. None D 5 €333,182: 2:232:53? “:3:le Frozen Dried Tomato Soup 4874 Vegetable Soup (incl. Minestrone Soup) 4991 Vegetable and Meat Soup 5580 Chicken Soup or Broth 5693 Mushroom Soup 4781 Pea Soup 4532 Other E1 Asparagus 1:1 Bean 4421 4702 1:] Celery D Onion 4131 4813 Check 1 mm PIES ...... 1:1 2:2"; 3:22: ..:::... Beef Pie 5597 Chicken Pie 5695 Turkey Pie 5621 Pizza Pie Mix 5581 Pizza Pie—Prepared or Frozen 5582 Check it DINNERS — Brought Home None D .82.:ng ordnh Anitf'Pbld Frozen Hot Chicken-in-the-Rough 0100 TN. Dinner. , 0100 Other (kind) 0100 ”2) PREPARED BABY FOOD (:0de None C] ____________.__ No. ol Welght Total No. ol Weight Total BREAD FRUITS 2:2: .522... “221:" VEGETABLES 2:2: ..:::. “221:“ m. Applesauce 3321—1— Beets 4710 WI Apricots 3321 Carrots 4470 i —— Bananas 3340 Green Beans 4440 White Non-Enric Peaches 3440 Peas 4530- m Pears 3450_ spinach 4260 ’ W Plums 3470 Squash 4560 E; Bread Prunes 3480 -Sweet Potatoes 4570 Cinnamon 3..., Other 3530 Other Vegetables 4900 l “pgigmd I jolt andllfali Mixed Vegetables 4990 Other DEV: x0230 Apple-Apricot 3601 Other Ii“; 1;: Pear-Pineapple 3602 “—K Other Mixed 3603— MEATS \ __ Beef 5150 _ m Chicken 5615 Biscuits Lamb 5250 ' ' Buns Apricot-Farina 3620 i—Liver 5143 ‘81? jun-#5310 “3021 . ' Other Meats 5552 m Apricot-Oatmeal 3622 \ 'Other mixed 3623— BreOktast I Mixed Meats 5551 7? er Roi \_ Dinner Ro' Apple Juice 3319 Eggs 5901 73m Orange Juice 3249 Eggs and Bacon 5902 m Orange-Apricot J. 3640 Soup (Baby Food) 4993 \, Orange-Prune J. 3641 . &E Orange-Apple J. 3642 T i V" I \EILME Pineapple Juice 34607—2 T VEG' 8‘ MEAT I \Ehiisf: Other Juice 3549 '_Veg. and Beef 5561 o w Veg. and Chicken 5662 C:Up Ca — Veg. and Bacon 5563 m Teething Biscuit;_w6_2_94_ Veg. and Liver 5564 . \F‘ —Pu-ddings 7421 Veg. and Lamb 5565 \fof ' Oiher 5566 l T 383% i $35 Baby Cereal (kind) ' \Cjiim ECIair: I l , : \ BAKERY AND CEREAL PRODUCTS (13) No .ot Lbs., 01., Total BREAD None D Loaves Each Loot Amt. Paid White Enriched Bread 6101 White Enriched Bread 6101 White Non-Enriched Bread 6102 Whole or Cracked Wheat Bread 6104 Rye Bread 6105 Raisin Bread 6106 Cinnamon Bread 6107 Diet Bread 61 12 Half and Half Bread 6117 Other 08 [3 Vienna, French 10 C] Date Nut 03 :1 Potato 61 09 C] Boston Brown 1 1 D Bran 03 [:1 Salt Rising Other 13 [1 Pumpernickel 15 a Rusk 61 14 C] Zwieback 16 C] Toast C) ......... No. in Lbs., 02., Total Amt. BUNSI ROLLS, CAKES N°"° E] No. Pkgs. Each Pkg. Each Pkg. Paid BISCUITS 1 C] Oven Ready 2 D Refrig. (can or tube) [:1 ..... O20 Buns 1 D Hamburg 2 C] WTener D ............... 621 Buns 1 (:1 Hamburg 2 C] Wiener D ................ 621 1 [:1 Rais'n 3 C] N t Co e ed 5 [:1 Pla'n Frosted Breakfas’ ROMS 2 I] Cinn|amon 4 C] Britterscvotrch [:1 . . J ......... 622 1C1R"n 3DNtCoered 5DPI'F td Breckfos' ROIIS 2 [:1 Cciiihlamon 4 [Z] Buutterscvotch D . .0.": . 5:31: . 622 . 1 [j P rkerho se Dlnner Rolls 2 C] ngpy Sged 3 D Brown and serve 623 - 1 C] Parke ho e Dmner ROHS 2 Cl Poppyr Seudd C) ............ 623 Other Rolls 1 1:] Whole Wheat El ................ 624 IDPII 3DEgI'sh 5C1Corn MUffins 2 D 8:"? 4 C1 BTueberry ............ 625 Cofl‘ee Cake 6260 Angel Food Cake 1 1:1 Iced 2 1:] Not Iced 630 Cheese Cake 6310 Layer Cake 1 [3 Plain Frosling 2 D Choc. Frosting 632 I D Chocol t , F t d 3 [:1 Chocol t Not Fro ted Cup Cakes 2 1:] Plain, £065,030: e 4 C] Plain NooteFrosted s 633 Fruit Cake 6340 100* Cake is B 33.23333... 3 D Fif’f‘3'?.'?f'f’?'fii ..... 635 Sponge Cake 1 E] No Frosting 2 C] With Frosting O36 Jelly Roll 6370 Cream PUFFS 6375 Edans 6378 “4’ BAKERY AND CEREAL PRODUCTS (cont.) COOK|ES None I] th., 02., Total Amt. No. Pkgs. Each Pkg. Pald Cookies—Plain or sugared 6410 Cookies—Plain with Frosting or Marshmallow 641 1 Cookies—Chocolate 6412 Cookies—Chocolate with Frosting or Marshmallow 6413 Cookies—Fruit Filled 6414 Cookies— Assorted 6415 No. In Lbs., 01., Total Amt. DOUGHNUTS’ PIES None D No. Pkgs. Each Pkg. Each Pkg. Paid - IDFrotedorSg red DoughnUIs—chsed 2 D Nots Frosted :rOSugared 642 3 D Frosted or S gored Doughnuts—Cake Type 4 D Not Frosted I6r Sugared 642 Doughnuts—Bismarks 6426 - 1 D Cream 3D M nceme t 5 D Ch'tton FIGS 2 D Custard 4D P ulmpk' kin O . .l. ........... 650 - - 10 D Apple 12 D Bl ckberr 4D Boysenberry PleS—FTUII I 1 D Apricot 13 D BISeberrny 5D Cherry 65 - - 16DPeach 18DR "n 2ODRh barb P1es—Fru1t 17 D Pineapple 19 D Rgispberry 21 D Strtbwberry 65 . - 22 D Apple-Bl berr 24 D Str wberr -Rh barb PIeS—Frmt 23 D Apple-Chfrry Y D . ..0 ..... .y. . .u. ..... 65 None Lbs., Oz., Total Amt. MIXES D No. Pkgs. Each Pkg. Paid . 1D Wh te 3 D Chocol to 5 D Sp'ce Cake M'x 2D Yellbw 4 1:] Angel Food ..I ......... 66] . 1D White 3 D Chocol to 5 D Sp'ce Cake MIX 2D Yellow 4 D Angel Food D ...I ........... 66] . . 1D Pla' n COOk'e MIX 2:] Chdcolate D ................ 662 Hot Roll or Biscuit Mix ; B :33?" 663 - 1 D Corn 3 D Plain Bread MIX 2 CI Bran D ............... 664 . 1 I] Corn 3 D P10 11 MUffin MIX 2 D Bran D . . .I ............ 665 Pie Crust Mix 6660 Complete Pie Filling and Crust Mix 6661 Pancake or Waffle Mix igiiéhwh‘i‘t'i ................. 667 Other Mix (kind) 668 Lbs., Oz., Total Amt. FLOUR, CORN MEAL None D No. Pkgs. Each Pkg. Paid Cake Flour 6690 Graham Flour 6691 - 2 D Enr'ched White Flour 3 Cl NOI‘-En,iched 669 Corn Meal 6694 Corn Starch 6695 Other Flour 99 [1 Whole Wheat 87 [1 Barley 66 97 D Buckwheat 96 D Rye BAKE CIACKERS O I 33:: Crackers, others . C39? 50 led Crackers {cram Crockeis __— Otter Crackers Ilide / Crater Meal ___,_______ _______————— —— N. SPAGHEIII E' I I I “— Sccghetti or Mccc Spaghetti 01 Mocr \ N'codles N Rice Minute RICE \ Tapioca \ \ BREAKF A81 \ \\ Reody-to-Seri Reed 4 . I° Be Cooke \“ Io Be CC‘OIIG \‘ I \ \ \ Al’PE'rIz \ Cfcckeqci \\‘ I CheeSe TV \\ PCPDed ( P\ PTQIIeIS I 332833 BAKERY AND CEREAL PRODUCTS (cont.) “51 CRACKERS None D Lbs., 02., Total Amt. No. Pkgs. Each Pkg. Paid Soda Crackers, Others Similar 6280 Other Salted Crackers 6281 Graham Crackers ; B 322°31‘22" £333 629 Other Crackers (kind) 6295 Cracker Meal 6296 Check One I Check One Here Too No". D I x 33 4 8 6 “It: 3 mg . 2 Lbs., 01., SPAGHET", ETC. c . 5 3 :3 a E g No. Pkgs. Each Pkg. Total 3 a 3 2 ° 3 ° .2 o '5- or Cans or Can Amt. Pald at U a: a 5 O S u l- O Spaghetti or Macaroni 67 Spaghetti or Macaroni 67 Noodles 68 Rice 69 Minute Rice 6917 Tapioca 6959 lb.,0., TtlA t. BREAKFAST CEREAL N... [:1 ,0 .... W 50;” Pk; . gown Ready-to-Serve (kind—Wheaties, etc.) Ready-to-Serve (kind) Ready-to-Serve (kind) To Be Cooked (kind) To Be Cooked (kind) APPETIZERs, ETC. N... [j .9 Crackeriacks, Karmel Corn 6970 Cheese Twist, Cheese Sticks, Corn Pone, Etc. 6971 Popped Corn, Fritos 6972 Pop Corn—Raw 6973 Pretzels 6974 Cones for Ice Cream 6977 Other (kind) OTHER GRAIN PRODUCTS None D (161 SUGAR, SWEETS, CANDY SUGAR NONE D 7100 Number 0! Pounds . ( Prlco per Pound Total Amt. Pold White or Powdered 7100 Dextrose, Maltose 71 10 Brown 7120 Maple 71 30 ls-lBRhEEJEPY AND NONE D 7200 "2.353.1'31’15' Pm. per Unit Total Amt. Paid Corn Syrup 7210 Cane Syrup 7220 Maple Syrup 7230 Molasses 7240 Sorghum 7250 Other Syrup 7260 Toppings for Ice Cream 7265 ‘Honey 7270 CANDY 8‘ SWEETS NONE E) 7300 "02:: 5.1.1:?“ Pm. per Pound Total Amt. Paid Fondant, Mints, Marshmallow 7310 Cream—Chocolate Covered 7320 Cream—Fruit, Nut 7330 Cream—Chocolate, Fruit, Nut 7340 7— Solid Chocolate 7350 Other Candy or Sweets 7360 REPAR D EESSERTE MIXES NONE a mo 12:122.? 23:12: T... ...... p... Gelatin— Unfiavored 7410 — Flavored Gelatin (Jello, Royal, etc.) 7415 Puddings or Pie Fillings—Dry Mix 7420 — Candy Mix 7440 — Frosting Mix 7450 NUTS AND NUT PRODUCTS Check Ono fl NONE [3 35-37 11:32:! ”:32...“ “13:.” In sh." Shelled Coconuts 751 Peanuts 752 Peanut Butter 7532 Almonds 754 Brazils 755 Walnuts, Pecans 756 Mixed Nuts 757 Other Nuts (kind) 758 Have you included all of the food purchases by other members of the household? Do not include sales tax in reporting price or total amount paid. and Total Amt. Paid m I 1 Total“ ’ In Sh!" ! smliod AMI-'0 ‘/1 BEVERAGES (17) Size of UnIt NONE D 8000 B::1T::t::'e. 02533.11?” etc. prrflljzit Am:::i‘tllPaid Beer 81 10 Liquors (Whiskey, Gin, Rum, etc.) 8220 Wine 8230 Cocoa 0 [I] Mix 1 [:1 Plain 831 CofTee—Ground 8420 Coffee—Instant 8421 Tea 1) B (3:00.; 2 [3 Instant 853 Soft Drinks—bottled 8610 Soft Drinks—Liquor Mixes 861 1 Soft Drinks—powdered 8720 Postum 8810 Ovaltine 881 1 Other (kind) NONECI VITAMINS AND MINERALS VITAMINS (name kind) 9100 Quantity Purchased Total Amount Paid MINERALS (name kind) 9200 COOKING AIDS Number lbs., 02., Price Total NONE D 9300 at Unite Each Unit per Unit Amount Paid Baking Powder 931 1 Baking Soda 9312 Canning Aids (Certo, etc.) 9369 Chocolate—Baking, Chocolate Chips 9321 Extracts (name kind) 9330 Junket Tablets 9345 Mustard—Prepared 9436 Meat Sauces (name kind) 9340 Meat Tenderlzer 9375 Salt 9350 Spices (name kind) 9360 #Pepper 9361 Vinegar 9371 Yeast 9372 Cream of Tartar 9374 (18) Directions for Keeping Your Diary Foods to be reported You are to report all food items including beverages brought into your home whether (I) purchased, (2) received as a gift, (3) home-grown, or (4) obtained by hunting and fishing. Items that you buy or grow and give to someone as a gift are not to be reported in your diary. Meals purchased and eaten away from home should be reported as follows: On page 19 of your diary report the number and total amount spent for such meals away from home. If members of your household are guests for meals ' away from home report the number of such meals. If meals eaten out are paid for by your employer they are reported similar to guest meals, that is, report the number of such meals, but not the amount spent for them. If you serve meals to guests enter the guest meals on page 19. Do not report meals that you serve to guests in a restaurant if you pay for the meals. Do not report in your diary items such as Coca-Cola, ice cream cones, candy bars, etc. that are purchased and eaten away from home unless you consider them a complete or regular meal. Food that is brought into your home and later used for school lunches, workmen's lunches, picnics, etc. should be reported in your diary. Meals comprised of such foods are not cansidered meals away from home. When to repart items in your diary Please record food which you purchase on the day it is purchased. ReCOrd _gr_f£ items on the day they are received. Record home-grown foods on the day they are br0ught into the home (or put in your frozen food locker if you rent one). Record items obtained by hunting and fishing on the day they are brought into the home (or put in your frozen food locker if you rent one). Please be careful not to report any food items twice. For example, do not report home-canned items both when they are purchased or brought into the home and again when they are used. Products which you had on hand before you became a panel member are not to be recorded. Special points on reporting Products are prepackaged if they are wrapped or placed in a consumer size package with a specific price indicated for each package before the customer enters the store. The opposite is “bulk" in which the product is not wrapped or placed in a consumer size package bolero the Customer enters the store. Frozen products are frozen when you buy them. Do not check as frozen the unfrozen items you buy to take home and later freeze. If the product is frozen when you obtain it, check frozen regardless of the type of container it is in. In reporting fruits and vegetables please indicate the amount in weight or liquid measure, except for those Mfruits and vegetables indicated with “qts., no., heads, ears, or bunches" after the name of the product it is all right to express the quantity in these units. For example, fresh strawberries may be recorded as number of quarts, fresh grapefruit may be recorded as 3 grapefruit or I doz. grapefruit, etc., head lettuce may be reported as number of heads, or radishes may be reported as number of bunches, etc. Unless one of these units follows the name of the product, please report fresh products in paunds and ounces, or if bought in large amounts pecks, bushels, etc. may be used. All frozen, canned, dried, iam, ielly, and soup items should be recorded in weight or liquid measure. If you are not sure of the weight of home-grown, gift, or hunting and fishing items please make an estimate and enter it in the diary. Vacations, trips, etc—if you are going to be away from home for 2 months or less please arrange for a supply of diaries to lost during the period. Change in household membership—In reporting a change in the size of household report only changes which you expect to last 2 months or more. If less than 2 months, report visitors’ meals as guest meals and meals of household members who are away as meals away from home. Reporting income—Record the income of members of your household each week that earnings in any amount are received. All of your income should be reported, except that federal income tax preferably should be deducted. No other deductions from income (i.e., car payments, social security, bonds, retirement, etc.) should be made when reporting. Do not report sales tax as a part of the expenditure for food items. Do not include bottle deposit as part of the cost of mill: and beverages. EXTRA SPACE (for items not listed in diary) ”9’ Number Size Price Total Description of Units of Unit per Unit Amount Paid VITAL DATA QUESTIONS 1. Has there been any Change in your household membership since your last reporting week? YES NO (Circle one) If yes, what was their: Relationship to homemaker Age How many are there in your household now? 2. How many regular meals were eaten away from home by members of your household last week?________-lOne meal consists of either breakfast, dinner or supper for ONE person). Total amount spent 3. How many guest meals were served during the past week? (A guest is anyone not a regular household member). 4. What was the total income payment actually received during the diary week by: The male and female head of the household? Other members of the household? Check if none 5. Was this before or after Federal Income Tax deductions? Before ( ) After I ) (In reporting income payments, please keep in mind that they might come from many sources. These include wages, salaries, commissions, pensions, interest and dividends, annuities, profit from business and professional services, profit from rent, government payments, gifts, and any other sources.) This information will be held strictly confidential, and your name will not be associated with it. It is necessary to ask these questions in order to get the greatest value from your diary. Professor G. G. Quackenbush M. S. U. Consumer Panel Michigan State University East Lansing, Michigan (osoH do]; powwng ossod) NO POSTAGE STAMP NECESSARY POSTAGE HAS BEEN PREPAID BY Professor Gerald Quackenbush M. S. U. Consumer Panel Michigan State University East Lansing, Michigan _ memes / MICHIGAN STATE UNIVFRSTTY ,/ EAST LANSING MfCH ..4.::"«.2m 10.19 v———-— — . #— /’30 ‘ T tfif’g I r . ' 0 Q“: ‘/V t/ v (r 56L 4 3150 9 II 3 i I 1 ll 3 l ~—V"r' MICHIGAN STATE UNIVERSITY LIBRARIES if"; _ ' . i 9' I <3; , . ..:.-'3, 1‘! s . ...: 1 Li. 5 ' --:: t I a...“ c g. Q] psi—I LL .5": j