Food Prices: An Overview of Current Evidence

Listen

6. Consumer Response to Food Price Increases

6.1 The Income Effect

How the demand for goods is affected by changes in prices is influenced by many factors. If the product is a vital component in the household diet, the level of its consumption is unlikely to be affected significantly by changes in its price. This is usually the case for necessities which have few if any alternatives. Generally food is a necessity, so when all food prices rise uniformly, the consumer is limited in the extend to which they can change their behaviour. Faced with higher cost of food, consumer can either cut down on their levels of food consumption or substitute products within the food category.

Figure 11 shows that the number of food items purchased per household on a 4 weekly basis has remained relatively stable over the year to December 2008 despite large price increases. This should be expected, given that people need a certain intake of food to function properly. However, according to WRAP, Scottish households waste 61% of food that could have been eaten 29, thus indicating that more food than necessary is being bought. It is surprising that consumers do not seem to have responded to higher prices by cutting down on waste.

Figure 11 Spend per household and Packs per household in Scotland Dec 2007 - Dec 2008

Figure 11 Spend per household and Packs per household in Scotland Dec 2007 - Dec 2008

Source: TNS Worldpanel

6.2 The Substitution Effect

Consumers, however, have responded by substituting within the general category of food and drink by switching to cheaper produce. This has taken form of both switching to a different type of retailer and trading down food quality categories. Figure 12 shows that discounters such as Lidl, Aldi and Iceland increased their market share in Scotland by increasing their growth rates from -1.5% during 12 week period ending 2nd December 2007 to 22.8% in 12 week period ending 28th Dec 2008. 30 Premium retailers such as Waitrose and M&S suffered most as their growth rates plummeted from 18.2% in 12 week period ending 2nd December 2007 to -3.6% in 12 week period ending 28th December 2008. Although much variation could be observed amongst mainstream retailers such as Tesco, Somerfield, Morrisons and Asda, together their sales growth rates have declined from 9.4% to 1.4% during the same period, whilst Sainsbury's growth rate has remained relatively stable.

Figure 12 Growth in Supermarket sales by Price Banding in Scotland Dec 2007-Dec 2008

Figure 12 Growth in Supermarket sales by Price Banding in Scotland Dec 2007-Dec 2008

Source: TNS Worldpanel

Figure 13 shows that Scottish consumers have also been substituting goods according to their label. Organic own label products were growing at the highest rates of 23.4% at the start of 2008 but their growth rates fell to -7.5% by October 2008. Premium Own Label goods have also been on a downward trend during 2008, with standard own label sales growth has remained stable. The only category that has expanded significantly was Value Own Label. This category was showing negative growth at the beginning of 2008 but increased significantly during the year to 39.5% in the 12 week period to October 2008 reflecting consumers reaction to increased cost of living.

Figure 13 Growth in Supermarket sales by own label product category in Scotland Dec 2007 - Dec 2008

Figure 13 Growth in Supermarket sales by own label product category in Scotland Dec 2007 - Dec 2008

Source: TNS Worldpanel

All of the above data point to the fact that the income effect of food price increases has so far been relatively small as there is no evidence of people buying less categories of food produce, which indicates that that the quantity of food purchased is roughly the same. However, the substitution effect within food is large, as evidenced by growth in discount retailers sales and economy range products and the simultaneous decline in sales of premium retailers and foods.

This effect is due to the great availability of substitutes in the food and drink retail market, with similar goods available at different prices. So far it looks like the consumers have not felt the necessity to cut down on their total consumption of food and instead have opted for trading down labels and retailers. However, it is important to remember that whilst these figures paint a good picture of consumption patterns for the whole of Scotland, they fail to capture the effect on individual groups of people. The low income groups, who before the price rises were already purchasing their food from discount supermarkets and consumed mainly economy based ranges may have struggled to trade any further down and have possibly had to cut back on their food consumption which may have had consequences on their health and well-being.

The low income consumers may also have experienced a higher rate of inflation due to their expenditure structure being different to that of those on high incomes. Comparing the poorest and the richest deciles, shows that food purchases constitutes 17.9% of the total expenditure of the poorest decile and only 9.4% of the total expenditure of the richest decile. Figure 14 shows the inflation trend for Scotland based on CPI prices and expenditure shares from Expenditure and Food Survey 2006. 31 The graph shows that in the recent months, the poorest in the society experienced higher inflation than the richest. This is due to the fact that the goods that the poorest spend most of their income on increased in prices significantly during that time, with food prices being the main contributor to that trend.

Figure 14 - Inflation in Scotland - Poorest (1st) vs Richest (10th) deciles

Figure 14 - Inflation in Scotland - Poorest (1st) vs Richest (10th) deciles

Source: Scottish Agricultural College

6.3 Food demand price elasticities

Whilst there is no evidence to imply that total consumption of food has decreased as a result of food price increases and Figure 11 seems to suggest that the spend and the volume of food purchased in Scotland has remained stable over the last year, it is very likely that the composition of that expenditure has changed in response to changes in prices. Trends outlined in Section 6.2 show that Scottish consumer have switched between retailers and product labels in an attempt to reduce their food expenditure. However, it is likely that consumers have also changed the composition of their shopping basket in terms of food categories they purchase. If this is true, then it can be concluded that food prices have a major impact on household diet and therefore longer term impact on their health and well-being. This section looks at how consumers react to changes in prices of individual products by looking at price elasticities of a selection of foods for the UK and Scotland.

The responsiveness of demand to changes in the price of a good is referred to as demand price elasticity. Price elasticity looks at how the quantity of a good purchased changes as its price changes and measures the ratio of proportional change in quantity demanded of the good to the proportional change in price that brought it about. Unit elastic demand (absolute value of price elasticity = 1) means that the total expenditure on the product is not affected by the change in prices as the demand changes at the same rate as the price. The closer the price elasticity is to zero, the less the quantity consumed is affected by the price changes.

The more substitutes a good has, the more elastic the demand for it is going to be. If the price of that good increases, consumers can just switch to other products. The same works in the other direction. If the price of a good falls, the effect on demand will be larger if it has a lot of substitutes - the magnitude of the switch will be larger. How essential the product is in a household basket is also a factor. For luxury goods, elasticities are usually lower in absolute terms - if the price increases, households can reduce or forgo consumption of that good completely. As a whole, food is a necessity - households need to consume a certain amount of food. Many food products are not essential - an example of that are confectionary and alcohol. However, non-essential or luxury goods could be habit-forming with very low elasticity of demand.

It is important to distinguish between short term and long term elasticities of demand. In the long term, consumers have more time to change their behaviour in response to price changes, therefore the demand is likely to be more elastic. This is because consumers will sometimes take time to discover alternatives to the product if its price increases or they may have already invested in a complementary durable good (e.g. coffee maker, petrol car).

Table 2 below shows the price elasticities for various foods in a consumer basket along with the budget shares of each from the National Food Survey 2000 conducted by the Office of National Statistics. Monthly changes in prices and quantities purchased were used to derive the elasticities. The results indicate that fresh fish, sugar and other fruit and fruit products are sensitive to changes in their price. The demand for other cereals and cereal products is almost unit elastic. On the other hand, the demand for fresh potatoes, milk and processed and shell fish and prepared fish is least sensitive in responding to price changes. Cheese, eggs, fresh vegetables and fruit (including bananas) and frozen fish are also insensitive to price changes which indicates their importance in the consumer basket.

Table 2 Own price elasticities of food and drink products, 1988-2000

Table 2 Own price elasticities of food and drink products, 1988-2000

Source: National Food Survey 2000, Office of National Statistics

Analysis of multiple retailer data on price changes and quantities purchased over the period of October 2006 to September 2008 allowed calculation of own and cross price elasticities of 12 products. The results show that only white bread, fresh new potatoes, whole chicken and minced beef are price elastic (i.e. with elasticities of above 1 in absolute values). This means that as the price on those products goes up, the quantities of them purchased fall. The elasticities of the rest of the products examined were found to be less than 1 in absolute values, making them price inelastic.

Table 3 Estimated demand elasticities for the selected products

Table 3 Estimated demand elasticities for the selected products

Source: Scottish Agricultural College

The above elasticities can be used in conjunction with price inflation on individual products given in Figure 15 to establish what effect the recent price rises may have had on the consumer expenditure on individual products. This task is complicated by the fact that there are differences in the way different products are grouped together. Despite this some implications of the recent price increases can still be presumed. For example, according to RPI data in Figure 15, beef and eggs have seen the very large price increases in the 18 months to February 2009. However, the price elasticity of demand for eggs is very low according to both ONS and SAC results in Table 2 and Table 3, which implies that despite price increases, the consumer may have only decreased their consumption of eggs only marginally, if at all. However, the price elasticity of demand for minced beef is much higher in absolute terms and the consumer response to price rises is different in that the consumption of it is reduced greatly. This is more likely to be due to availability of substitutes to minced beef in the form of other meats.

Figure 15 - Cumulative Retail Price Inflation on Different Food Categories - September 2007 - February 2009

Figure 15 - Cumulative Retail Price Inflation on Different Food Categories - September 2007 - February 2009

Source: Office of National Statistics

Price elasticities of demand help us determine how changes in prices affect consumer purchasing power. Interestingly, increases in skimmed and semi-skimmed milk and fish prices, which have been smaller than those of, for example, chicken (20% price increases on poultry since September 2007 and 11% and 12% increases on fish and milk), would have had a greater effect at reducing consumer purchasing power, as consumers seem to find it harder to substitute for those goods. This is reflected in the differences in elasticities of those goods.

6.4 Distributional impact - by income level

Food security is of high importance at the national level, however, even if a country is food secure in wider terms, it does not necessarily mean that that is the case at the household and individual level. Equal access to food is an important aspect of food security and if prices are too high for people on low incomes to buy enough food, such people should be considered as food insecure.

The average household in the UK allocated 9% of its expenditure to food. However, those on lower incomes spend proportionately more compared to those on higher incomes. This is evident from Figure 16, which shows how percentage expenditure on food falls as income increases. In 2005/2006, the poorest UK households spent 15% of their income on food and non-alcoholic drinks, whilst the richest spent 7%. 32 These figures mean that increasing food prices essentially redistribute away from people on low incomes.

Figure 16 Share of household expenditure spent on food and non-alcoholic beverages 2005/2006

Figure 16 Share of household expenditure spent on food and non-alcoholic beverages 2005/2006

Source: Defra, ONS

Considering the above, rising food prices have a disproportionate effect on low income groups who spend a greater share of their budgets on food.

There is evidence that insufficient income negatively affects UK households consumption of food. According to a survey by the Food Standards Agency 33, among those who did not always have the kind of food they wanted, 79% stated not having enough money as the reason and 29% of the population on low income stated that their access to enough food had been limited by factors such as lack of money or other resources (e.g. storage facilities, transport) at some time during the previous year.

In addition, according to the same survey 39% of the low income population reported that, in the last year, they had been worried that their food would run out before they got money for more, while 36% indicated that they could not afford to eat balanced meals. Overall, 22% reported reducing or skipping meals, and 5% reported not eating for a whole day, because they did not have enough money to buy food.

Figure 17 Percentage of adults scoring positively on survey questions

Figure 17 Percentage of adults scoring positively on survey questions

Source: Food Standards Agency

These figures is based on the results of a survey conducted during the period between November 2003 and January 2005. Given that food prices have increased significantly over the last year and the income levels have not done so on the same scale, the problem of food insecurity of low income households could only have been exacerbated.

Given that food is purchased by everyone, the rise in their prices makes everyone worse off in terms of income available to buy other goods. Of course people may also cut back on their consumption of food but there is a limit to the extent to which they can do that without putting their health and well-being at risk. Overall, the amount of goods and services people can buy will be decreased as the result of rising food prices.

An important aspect of that is how different groups will be affected - the distributional impact of rising prices. Given the above statistics, there is a part of population of the UK who could be severely affected by the increases in prices. Already on low income, these groups could have been hit most as the cost of living went up, meaning there is even less food they can afford or the quality of food they buy is significantly lower.

Economic theory links consumption to utility, i.e. the more is consumed - the higher the utility. If theory of diminishing marginal utility is applied, those on higher incomes would suffer a smaller loss of utility from decreased consumption than those on lower incomes. In other words, as income grows, the satisfaction derived from an additional unit of consumption declines. This implies that the rise in food prices may also have a large distributional effect in terms of utility. The social welfare function takes this concept into account by linking personal utility (level of satisfaction) to income 34 and weighing the utility of those on lower incomes higher than the utility of those on higher incomes. Thus the overall negative impact of rising prices on necessities such as food could be very large.

6.5 Distributional impact - by geographical disparities

A study undertaken by the University of Edinburgh for the Food Standards Agency (Scotland) 35 involved identification of 9 survey areas, termed survey sentinels, in which detailed surveys were undertaken at all shops within the area. To select the sentinels the data zones within the six-fold Urban-Rural classification were divided into three types of environment and then stratified into deciles of relative deprivation ( SIMD).

Figure 18 Sentinel sites used in Accessing Healthy Food study

Urban/rural type

Relative Deprivation

Sentinel code

Sentinel name

Median SIMD decile

Interquartile SIMD range

Urban

Deprived

UR2

Scotstoun/Drumchapel

9

8 to 10

Urban

Affluent

UR1

Broughty Ferry

2

2 to 4

Urban

Mixed

UR3

Inverness

4

3 to 7

Small town

Deprived

ST1

Kilbirnie

7

5 to 8

Small town

Affluent

ST2e

Ellon

1

1 to 2

Small town

Affluent

ST2c

Cupar

2

1 to 4

Rural

Deprived

RD1

Dornoch

6

5 to 7

Rural

Affluent

RA1

Haddington

3

2 to 4

Island

Mixed

IS2

Orkney

4

3 to 5

Island

Mixed/deprived

IS1

Eilean Siar

7

6 to 7

Total

Source: University of Edinburgh

In each sentinel, all food shops were visited and the presence and prices of a range of healthy food products were recorded with the use of Healthy Eating Indicator Shopping Basket ( HEISB) - a list of foods which comprised a total of 35 items drawn from 5 major food groups (see Appendix). The survey sentinels were selected to represent different socio-economic environments, in respect of affluence-deprivation and urbanism-rurality. These data were collected for a total of 466 shops across the sentinels.

The two affluent small town sentinel sites were the cheapest places to purchase the total HEISB with rural deprived and island mixed/deprived being the two most expensive. The urban deprived study site fared relatively well and was cheaper or as cheap as rural and island sites irrespective of their deprivation status, and was also cheaper than the small town deprived site. In general price seem to rise with deprivation across the first four quintiles and fall in the fifth quintile.

This evidence shows that it is the most deprived part of the population that faced the highest prices on food. It has already been established that those on lower incomes spend more on food proportionately so it can be concluded that the most vulnerable areas of the Scotland are likely to have been affected most by the rising food prices.

6.6 Impact on diet

Scotland, as well as the UK as a whole is faced with the problem of unhealthy weight. In 2003, 43% of men and 34% of women were overweight and 22% and 26% respectively were obese 36. Excess weight is associated with over-consumption of food and poor diet and incurs high costs to individuals, their families and the society as a whole. Rising food prices may have a positive effect in terms of reducing over-consumption of food and thus cutting costs associated with unhealthy body weight and related diseases (see Section 7).

However, if price rises are not uniform across different food products, the problem of unhealthy eating can be exacerbated even further. Food products vital for healthy diet such as fruit and vegetables, are often expensive compared to other, less healthy alternatives, and not being fully aware of the benefits of their consumption many people choose to buy cheaper produce in larger quantities.

In the current climate, retailer in-store promotions have an important influence on people's food purchases. A survey by the National Consumer Council conducted in Sheffield counted over 4,300 price promotions over the eight stores surveyed, which constitutes an increase of over 17% since 2006. 37 With food prices high on the agenda, retailers are focusing even more on special offers, prompting customers into multiple purchases at discounted prices. The survey also shows that fatty and sugary foods accounted for over half of all price promotions in the UK's leading retailers. This indicates that rising food prices may have the effect of skewing consumption towards unhealthier choices. If this is a true reflection of reality, deteriorating health of the population could be an additional cost that is not being taken into account.

Another study by National Consumer Council 38 claims that low-income consumers are being short-changed on health, as they buy more of the economy ranges own label products. The study found that many of the economy range foods contain more salt, fat and sugar than their standard equivalents.

Figure 13 shows that there has been a great shift towards economy range goods in the recent months, which implies that consumers are increasingly buying foods that are of lower nutritional quality. If this trend continues, the health impact could be significant.

Price promotions often involve below cost selling - this is when an item is sold for less than its input cost. Regulation can be imposed to prohibit below cost selling but the effect this would have on competition is unclear. Below cost selling undermines competition, especially smaller retailers 39 but restrictions on it set a price floor which can also limit competition and increase stock-management costs 40. Also, regulation against price promotion would also prevent healthy produce such as fruit and vegetables to be sold at lower prices, putting those trying to maintain a healthy diet on a budget at a disadvantage.