Effects of Milk Solids Content on Consumption of Milk R. D. B O Y N T O N Dairy Institute of California 1127 11th Street, #718 Sacramento 95814
Strong interest continues for higher solids-not-fat standards for whole, lowfat, and skim milk than those currently established by the F o o d and Drug Administration for milk in interstate commerce. This paper traces the implications of establishing new national fluid milk standards identical to the current high standards in California. The greatest change in product composition implied by the adoption of the California standards would be in lowfat milk, where the solids-not-fat standard would rise from 8.25 to 10.0%. Whole and skim milks' composition would be less dramatically affected. Fortification costs are estimated at 3~/.5 gal (1.89 L) for whole milk, about 55 for lowfat milks, and about 1.5~ for skim milk. Retail prices are projected to rise 3 to 5% across the fluid product line. Assuming no change in consumer preferences, fluid milk sales are estimated to drop 1.2 to 5.1% nationwide upon adoption of the California solids standards (the percentage drop dependent upon the responsiveness of consumers to price changes for fluid milk). Total sales of solids-not-fat would increase 21 to 184 million ib (19.52 to 83.46 million kg) annually under higher solids standards, which is not enough to affect significantly the current supply-demand imbalance in the dairy industry. INTRODUCTION
Strong interest continues particularly among dairy farm groups, for standards of identity with higher nonfat solids in fluid milk products.
Received August 26, 1985.
1986 J Dairy Sci 69:1454--1461
Legislative proposals have been introduced in the US Congress, a petition has been filed with the F o o d and Drug Administration (FDA) by the National Milk Producers Federation (NMPF), and a few state legislatures have recently addressed higher fluid product solids standards. The issue is complex and closely related to multiple component pricing and C o m m o d i t y Credit Corporation's purchase activities under the dairy price support program. Proposals advanced for higher solids standards vary slightly, but all call for higher solids-not-fat (SNF) standards for whole, lowfat, and skim milk than those currently established by F D A for milk in interstate commerce and higher than those established by all states except California. It is important to realize that the most significant change in standards being proposed is for lowfat milk products, where the minimum solids-not-fat would be increased over 20%. Milk fat in fluid milk products is the subject of much less concern, although there are some who feel that the whole milk standard should be set higher than the current FDA minimum of 3.25% and others who would prefer to see more uniform milk fat in lowfat milks. In 1962 California defined lowfat milk as having a milk fat content between 1.9 and 2.1 and 10% SNF. Prior to this time, lowfat milk was undefined in the state. These standards were based upon consumer preference research conducted by the University of California. Multiple component pricing for milk used in Class 1 products were developed along with the definition of lowfat milk. California's standards for fluid milk products are often used as the model for other states; in fact, a recently mandated Congressional study of the nationwide impacts of higher solids standards specified the California standards as those to be analyzed. This article is based upon that national study (with some updating) in which the author was the principal investigator (4).
SYMPOSIUM: IMPACTS ON THE CONSUMPTIONOF DAIRY PRODUCTS STANDARDS AND PRODUCT COMPOSITION
Table 1 shows current FDA and California standards for milk fat and SNF for whole, lowfat, and skim milks. Most state standards (applicable to intrastate sales) equal or closely approximate the FDA standards. It is for lowfat milk that California's SNF standards vary most from those of the FDA. Note that California's SNF standard for whole milk is not an absolute but rather can vary + .1% provided total solids equal 12.2%. This flexibility was added to the standard in 1977 at the request of processors who seasonally found minor fortification of whole milk necessary to meet the 8.7% fixed SNF standard. The California SNF standard for skim milk is .75% above the FDA standard; however, fortification of skim milk in California is uncommon. Incoming producer milk at 8.7% SNF, when standardized to .5% milk fat, has an SNF of about 9.0%. It is instructive to examine the milk fat and SNF content of fluid products currently marketed in the US. In general, fluid products are at or near the minimum federal standards. A 1982 federal order study in 15 central US marketing order areas provides a reasonably good indication of the content of fluid products being sold (5). Regular, whole milk averaged just under the m i n i m u m milk fat standard of 3.25% and averaged 8.56% SNF. Lowfat milks averaged well over the 8.25% SNF m i n i m u m but below 8.75%. Skim*milk averaged above the
TABLE 1. Existing Food and Drug Administration (FDA) and California milk fat (MF) and solids-not-fat (SNF) standards.
Fluid milk products
F DA California standards standards MF SNF MF SNF
Whole milk Lowfat milk Skim milk
3.25 .5-2.0 <.5
(%) 8.25 3.51 8.25 1.9-2.1 8.25 ~.25
8.72 10.0 9.0
I Allowed to equal 3.4 provided total solids are 12.2. 2Allowed to equal 8.6 provided total solids are 12.2.
FDA SNF m i n i m u m but below the California standard of 9%. This same Market Administrator study raised some concerns regarding enforcement of the present FDA solids standards. In their study of 72 handlers, 53% of all whole milk samples tested below 3.25% milk fat. Average SNF test of all unfortified fluid milk products tested was below the test of the producer milk from which they were made. The study also found that many products labeled as fortified had an SNF test no greater than that of the producer milk from which they were made. C O N S U M E R P R E F E R E N C E S A N D SALES OF F O R T I F I E D P R O D U C T S
Consumer preference for milk of varying solids composition has been the subject of considerable research. Most of this work, however, was done 10 yr or more ago. In the interim, research and casual observation suggest that tastes and preferences of American consumers have changed significantly. Most recent work is limited in scope. New consumer research is needed that examines the preferred composition of fluid milk products by different groups of consumers. Devero reviewed nine consumer preference studies conducted 1958 to 1966 (2) and summarized the findings of these works: 1) adding SNF improved the flavor and ac L ceptability of beverage milk; 2) additions of SNF were more easily detected than additions of milk fat, and SNF additions had the greater influence on consumer acceptance; 3) 2% fat, 10% SNF milk (2/10 milk) was invariably preferred to regular whole milk; and 4) fortified whole milks (3.5/9.5, 3.5/10.0, 4.0/9.5) were preferred to 2/10 milk. Recent informal studies by Associated Milk Producers, Inc. and the National Milk Producers Federation strongly suggest that a majority of milk drinkers prefer milk of higher SNF content than is currently offered in most markets. However, there is a threshold of added SNF below which most drinkers cannot discern differences. Dubensky (3) suggests that some consumers, especially skim milk drinkers, may prefer less SNF than those accustomed to higher total solids. Additional research is needed to cover several regions of the US and examine in detail the effects of personal characteristics and milk consumption habits on Journal of Dairy Science Vol. 69, No. 5, 1986
preferences for fortified products. This knowledge would be useful in designing solids standards to serve consumers, producers, and processors. Knowledge of consumer preferences is helpful in unraveling the solids standards issue but does not answer the most important question: would consumers buy more or less fluid milk if only fortified products were offered? What little research that exists on this subject is outdated, but it supports the idea that the higher price of fortified products discourages sales (6). Presently, fortified fluid products are offered for sale in most markets. In the last 15 yr, however, their share of the market has been steadily declining. Federal order data indicate that sales of fortified lowfat milk fell from about 77% of total lowfat milk sales in 1969 to about 15% in 1983. Fortified skim milk's share of skim sales fell from about 74% to about 26% in this same period. Numerous explanations have been advanced to explain this trend, but there is no general agreement on what has had the greatest impact. A frequently offered reason for the decline is that consumers have freely chosen unfortified products over fortified ones. Fortified products are higher in price, and some consumers likely do n o t judge their taste or nutritional attributes sufficiently better to justify the higher cost. The other principal reason offered for declining fortified product sales is that processors and retailers have gradually stopped offering SNFfortified products for sale as the focus of fluid milk competition has shifted to price and away from other milk attributes. In this regard, consumers have had little to say about this trend. Clearly, the chain stores, with their own store labels, have been a driving force in positioning milk this way. Chain stores saw little return from trying to promote fortified products particularly as their sales volume fell anyway in the face of changing consumer demand. As store brands commanded a greater share o f the dairy case, milk processors had little incentive to market aggressively the fortified products under their own brand. It seems clear that, at least to some extent, fortified products have been pushed out by the changing structure of food retailing and the related lack of consumer education about the value of fortified products. -
Journal of Dairy Science Vol. 69, No. 5, 1986
E S T I M A T E D COSTS OF F O R T I F I C A T I O N
Cost increases associated with higher solids standards have been estimated by many researchers. The estimates are quite consistent, although the baseline from which the quantity of additional necessary solids is measured does vary. In this analysis, the average milk fat and SNF tests of packaged products measured in the 1982 Market Administrator study were used as the baseline. If the test of packaged milk is below that of producer milk, the procedure used here assumes that the processor presently earns a return on the difference to be foregone if the standards were raised. Conversely, if the test of packaged milk is now above the test of producer milk, the baseline chosen for this analysis assumes that the processor's additional raw product cost with fortification accrues above the current tests of the packaged product. The difference between the California standards and the reported test of packaged milk products is used to measure additional components needed. Only for whole milk would the milk fat standard be changed. It is assumed that the existing mix of lowfat products (with fat tests ranging between .5 and 2.0%) would continue if the California standards were adopted nationwide; SNF standards would be raised to 10% for all lowfat products regardless of their fat content. In this analysis, the price of the added milk fat is set at 17 ¢/.10 ib (.045 kg) of milk fat, a value near the differential used in Federal Order markets. (This differential is slightly higher than the current one in most Federal Orders, but it does not materially affect the cost estimates obtained.) The SNF price is set at $.8914/lb (.454 kg) (a nonfat dry milk powder price of $.8575/lb (.454 kg) adjusted to a dry matter content of 96.2%). This price is up-todate with the current price of Grade A powder after the April 1, 1985 drop in the support price. Because added solids will displace skim milk, the savings from this displaced skim milk is accounted for at $.0899/lb (.454 k g ) b a s e d upon an average Class 1 price o f $14.63/cwt (45.4 kg) for milk of 3.5% milk fat and a 17¢ milk fat differential. In addition to the extra raw product costs associated with fortification to meet the California standards, there would be additional processing costs in the form of
SYMPOSIUM: IMPACTS ON THE CONSUMPTIONOF DAIRY PRODUCTS extra labor and the purchase of some necessary processing equipment. This additional processing cost is estimated at $.15/cwt (45.4 kg). As shown in Table 2, these assumptions result in net additional costs of fortification of $.015 to $.052/.5 gal (1.89 L), depending upon the particular fluid product in question. As expected, greatest cost increases occur for the unfortified lowfat milks. PROJECTED R E T A I L PRICE C H A N G E S
Based on the cost changes estimated in Table 2, two possible retail price change scenarios are presented in Table 3. Scenario 1 assumes that the full cost increases for each product are passed through the retailer to the consumer. This amounts to maintenance of present margins on each product. Scenario 2 assumes that processors/retailers would seek to maintain their present price differentials among the products in their fluid line by increasing the whole milk price by the full amount of the estimated cost increase for that product and then increasing the price of all other fluid products by an equal absolute amount. This scenario implies that margins would be maintained on whole milk, fall on most lowfat products and rise on skim milk. Obviously, neither of these scenarios is exactly what would happen if the California standards were adopted nationwide, but they represent reasonable
approximations of two feasible modified price structures. Clearly, competition at the wholesale and retail levels would determine the price structure that would ultimately emerge in each market. Estimated prices per half gallon (1.89 L) for six fluid products marketed nationwide under California standards are shown in Table 3, along with the percentage increase in retail price each implies. Under Scenario 1, retail prices increase from 1.5 to 4.9%, and under Scenario 2, prices rise from 2.8 to 3.1%. E S T I M A T E D EFFECTS ON C O N S U M P T I O N
The 1983 sales of whole, lowfat, and skim milks were established as the base against which sales changes would be estimated. Retail own price elasticity, a measure of the sales response to a change in the product's own price, greater than one (in absolute value) implies that the percentage change in sales exceeds the percentage change in price. Retail elasticities for whole, lowfat, and skim milks were obtained from Boehm and Babb (1). Elasticity estimates vary depending upon the estimation method used, the data set employed, and the time period studied. Consequently, no single elasticity estimate can be offered as best, but those of Boehm and Babb are considered sound. The own price elasticity of demand for fluid milk products is generally regarded as inelastic,
TABLE 2. Estimated cost increases for selected fluid milk products with adoption of California standards. Whole Reg1 Ingredient cost increases, $/cwt ($/45.4 kg) Solids-not-fat Milk fat Savings from displaced skim, $/cwt. ($/45.4 kg) Cost of additional processing, $/cwt. ($/45.4 kg) Net additional costs $/cwt ($45.4 kg) $/1.89 L) .5 gal
2% Lowfat Fort
1% Lowfat Reg Fort
I Reg = Regular, Fort = fortified. 2NA = Not applicable. Journal of Dairy Science Vol. 69, No. 5, 1986
14 5 8
although Boehm and Babb's long-run estimates are elastic. Both their short-run (SR) and long-run (LR) estimates are used in this analysis to provide a range within which consumer purchase response to higher prices should fall. These elasticities are 1) whole milk, - 1 . 7 0 LR, --.38 SR; lowfat milk, - 1 . 3 3 LR, --.55 SR; and skim milk, - 1 . 8 2 LR, - . 1 2 SR. These two sets of elasticities, in combination with the two possible price change scenarios previously described, yield four scenarios for which changes in fluid sales were estimated. Table 4 presents the estimates of reduced fluid product sales under each of the four scenarios for the six products affected by the adoption of the California standards. Note that this analysis assumes that fortification has no effect on consumers' tastes and preferences, that is, fortification goes undetected by consumers and the price increases discourage consumption to varying degrees. The upper half of Table 4 shows the estimated percentage change in sales and the lower half lists these changes in million pounds of packaged product. The greatest sales drops are forecast under the assumption of elastic
demand (LR). The "full cost pass through" scenario (no. 1) yielded larger sales reduction estimates than did Scenario 2 under either set of elasticities, because lowfat milk prices were increased the most under Scenario 1. For all fluid products taken together, under Scenario 1, US sales (excluding California) are estimated to fall 5.1% in the long run but only 1.5% in the short run. Under Scenario 2, aggregate fluid milk sales are projected to fall 4.4% in the LR and 1.2% in the SR. Table 5 shows the additional sales of SNF expected under adoption of the California standards, assuming that the new standards are exactly met. The projections represent both the change in the standards and the reduced volume of fluid products sold under the new standards. Solids-not-fat sales increased for all fluid products under all scenarios except for whole milk in the LR elasticity cases and skim milk in the LR under Scenario 2. In these cases, the modest increases in the SNF standards were offset by relatively large sales declines. The greatest additional SNF sales occur in the unfortified lowfat products, where standards were increased the most.
TABLE 3. Estimated increases in retail prices of selected fluid milk products with adoption of California standards. 1 Whole Reg2
2% Lowfat Fort
1% Lowfat Fort
($/.5 gal)3 Estimated retail prices under California standards Scenario 1 : Full cost pass through Scenario 2: Maintain retail price differentials Estimated percentage increases in retail prices Scenario 1 Scenario 2
(%) 2.8% 2.8%
1Average 1982 retail prices from "Dairy Outlook and Situation," September 1983, p. 7, Table 4, served as a baseline in this analysis. Given fairly stable retail prices in the past 3 yr, this period relates well to the present. Prices were not reported for individual lowfat and skim products. Consequently, the same reported lowfat (skim) price was used for all four lowfat products (both skim products). Subjective but reasonable assignment of retail prices for the various lowfat and skim products (consistent with the overall, reported average prices) resulted in almost identical percentage changes in retail prices. 2Reg = Regular, Fort = fortified. 3Dollars per 1.89 L. Journal of Dairy Science Vol. 69, No. 5, 1986
SYMPOSIUM: IMPACTS ON THE CONSUMPTION OF DAIRY PRODUCTS
TABLE 4. Estimates of reduced fluid product sales under four price and own price elasticity scenarios with the adoption of the california standards. Scenario 1, (full cost pass through) LR ]
Scenario 2, (maintain price differentials) LR
(% drop in sales) Whole milk Lowfat milks: 2% Regular 2% Fortified 1% Regular 1% Fortified Skim milk, regular
6.52 3.33 6.25 4.12 2.73
2.70 1..38 2.59 1.71 .18
3.72 3.72 3.72 3.72 5.64
1.54 1.54 1.54 1.54 .37
(million lb) 2 Whole milk Lowfat milks: 2% Regular 2% Fortified 1% Regular 1% Fortified Skim milk, regular Total fluid sales drop, million lb (.454 million kg) % Total US sales % All sales except California
701 48 169 37 55
290 20 70 15 4
400 53 100 33 113
166 22 42 14 7
2275 4.5 5.1
681 1.3 1.5
1964 3.9 4.4
533 1.1 1.2
1 LR = Long run, SR = short run. 2.454 Million kg; based on all states excluding California.
TABLE 5. Estimated changes in commercial sales of solids-not-fat (SNF) under four price change and own price elasticity scenarios with the adoption of the California standards. Scenario 1, (Full cost pass through) LR 1
Scenario 2, (maintain price differentials ) LR
(million lb) 2 Whole milk Lowfat milk 2% Regular 2% Fortified 1% Regular 1% Fortified Skim, regular Total increase in SNF sales
74 3 17 3 0
115 6 27 5 5
104 3 24 4 -5
127 6 29 6 4
] LR = Long run, SR = short run. 2.454 Million kg.
A n n u a l gains in SNF sales f r o m all fluid p r o d u c t s ranged f r o m a l o w o f 21 million lb (9.5 million kg) to a high o f 184 million p o u n d s (83.9 million kg). To p u t t h e s e n u m b e r s in perspective, t h e C o m m o d i t y Credit C o r p o r a t i o n had a l m o s t 1 billion lb (.454 billion kg) o f n o n f a t dry milk p o w d e r in storage o n May 31, 1985. Milk fat sales are e x p e c t e d to be o n l y m o d e s t l y a f f e c t e d with t h e a d o p t i o n o f t h e California standards. Because t h e fat s t a n d a r d is raised c o n s i d e r a b l y for w h o l e milk, milk fat sales in t h a t p r o d u c t are e s t i m a t e d t o increase. F o r all o t h e r fluid p r o d u c t s , fat sales are p r o j e c t e d to decline as sales v o l u m e declines. For all fluid p r o d u c t s t a k e n t o g e t h e r , fat sales are e s t i m a t e d to increase f r o m 11 million lb ( 4 . 9 9 million kg) (LR, Scenario 1) to 58 million lb (26.3 million kg) a n n u a l l y (SR, Scenario 2). As a p o i n t o f r e f e r e n c e , t h e g o v e r n m e n t held 163 million lb (73.9 million kg) o f b u t t e r in storage o n May 31, 1985. Journal of Dairy Science Vol. 69, No. 5, 1986
All of the scenarios examined to this point have assumed that retail prices for fluid milk products would rise and, implicitly, that there was no offsetting change in consumers' preferences for fortified products to prevent sales declines. It is possible to relax these assumptions and examine how fat and SNF sales would change with the adoption of the California standards. This case could represent either the situation where the fortified products were perceived by consumers to be sufficiently superior to the former products that consumers were willing to purchase the same volume at the higher price or the situation where retail fluid milk prices were not raised with the increase in solids standards but rather fat and skim milk prices to farmers were adjusted such that processors owed no more for raw milk after the adoption of the California standards than they did before. (This also assumes no additional processing costs.) In either case, changes in fat and SNF sales arise only from the incremental increases in the whole milk fat standard and the SNF standard for all six fluid products. Under this model, milk fat sales were estimated to increase by 73 million lb (33.11 million kg) annually and SNF sales would rise by 235 million lb (106.5 million kg) across all fluid products. Both these projections exceed the increases estimated under the previous four scenarios. CALIFORNIA'S CONSUMPTION EXPERIENCE
Fluid milk consumption in California is frequently cited as proof that higher solids standards will sell more milk. Unfortunately, no such conclusion can be drawn from the experience in California. An examination of 1980 per capita sales in California and the rest of the US shows that in all fluid product categories, California's sales exceeded the national average and the "all Federal Order" average. Total per capita consumption of fluid products in 1980 was 248 lb (112 kg) in California, 224 ib (101.6 kg) in the US, and 222 lb (100.6 kg) in all Federal Order markets. Lowfat sales in California were 80 lb per person in 1980 compared with national and Federal Order average of 72 lb (36.3 kg). Closer examination of the record, however, shows that other areas of the country had higher per capita sales in 1980 than California. An identical Journal of Dairy Science Vol. 69, No. 5, 1986
pattern emerges if one examines 1) absolute changes in per capita consumption or 2) percentage changes in per capita consumption during the 1970's: California's consumption record is better than the US average, but some regions of the country have outperformed California. Although higher solids standards might legitimately claim some of the credit for California's better than average per capita sales, there are at least four other factors that likely also played a significant role. These include 1) high expenditures on generic milk advertising in California for many years; 2) relatively low retail milk prices in California compared with other parts of the country; 3) a favorable demographic profile in California; and 4) the maintenance of especially high milk quality on the farm and throughout the distribution system in California. To date, no analysis has been undertaken that separates the effect of each of these factors on the level of milk consumption in California. Likely each played an important role. What is important to recognize is that the strength in California's consumption record cannot be attributed solely to higher solids standards. CONCLUSIONS
It seems clear that the erosion of solids amounts in fluid milk products over the past 20 years will not be reversed unless higher standards are mandated. Although it also seems clear that higher standards would improve product uniformity and restore some of the "quality" gradually lost while solids amounts were reduced to the minimum allowable, there are reasons to proceed slowly in efforts to raise the national standards. 1) Consumers' response is uncertain. Little is known regarding the product composition preferred by consumers or groups of consumers. Also, virtually nothing is known about how consumers might respond in the marketplace were solids standards to be raised. Extensive taste tests and market test research are needed to address these important questions. 2) The apparent inadequacy of the current standards enforcement system would suggest that improvements should be made there before higher standards are mandated. 3) The adoption o f higher standards for fluid products is inextricably linked to multiple component pricing o f producer milk. Unfortunately, much
SYMPOSIUM: IMPACTS ON THE CONSUMPTION OF DAIRY PRODUCTS remains to be learned a b o u t the optimal m e t h o d of c o m p o n e n t pricing and the m e t h o d s used to test accurately for m u l t i p l e c o m p o n e n t s . 4) National standards m a y n o t be the m o s t desirable; perhaps states or regions should have u n i q u e standards depending u p o n the tastes and preferences of these consumers. Only additional research will provide an answer to this question. 5) The host of available lowfat milk p r o d u c t s in markets outside California complicates t h e standards picture. S o m e hold that in the interest of p r o d u c t u n i f o r m i t y there should be only one, or at m o s t two, l o w f a t products sold. If this were m a n d a t e d , cost increases w o u l d be considerably higher than those estimated here because the fat c o n t e n t in m a n y lowfat products w o u l d have to be increased. This concern deserves m o r e a t t e n t i o n than it has received to date.
1 Boehm, W. T., and E. M. Babb. 1975. Household consumption of beverage milk products. Agric. Exp. Sm. Bull. 75, Purdue Univ. 2 Devero, J. E., 1973. Effects of varying nonfat solids and fat on consumer acceptance of milk beverages. J. Dairy Sci. 56:286. 3 Dubensky, M. S. 1983. Acceptability and stability of skim milk concentrated with reverse osmosis. Ph.D. diss., Dep. Food Sci., Cornell Univ. 4 Economic Research Service, US Department of Agrjqulture. 1984. Nationwide adoption of the California solids standards for fluid milk products: issues and impacts, Staff Rep. AGES840816. 5 Federal Milk Market Administrator Service Unit No. 1, Overland Park, KS. 1982. Composition of fluid milk products in selected Federal Order markets. 6 Hillman, J. S., J. W. Stull, and R. C. Angus. 1963. Consumer preference and acceptance for milk varying in fat and solids-not-fat. Tech. Bull. 152, Agric. Exp. Sin., Univ. Arizona.
Journal of Dairy Science Vol. 69, No. 5, 1986