The prices we pay: Why they matter and where they come from
The extent to which people shop around for the best price for products varies by income level. Households on relatively lower incomes tend to spend more time hunting for bargains and often choose budget options for products where possible. This column presents new evidence using consumer data from the United States, highlighting how prices differ across income groups. It shows how retailers’ response to households’ shopping behaviour reduces inequalities and provides policy-makers with useful lessons for supporting lower-income households.
Most media coverage of inequality tends to include the latest numbers on the distribution of income and wealth. Policy-makers are concerned about differences in people’s personal incomes and assets because having less of either means having less money in the pocket to spend. And with less money in your pocket, you cannot afford to rent a nice apartment, buy organic vegetables, or upgrade to the newest phone. In short, less money to spend means a lower standard of living.
Unfortunately, looking at inequality in spending alone can be misleading. The reason is that how much an extra euro of spending increases a person’s living standards depends on the price that they pay when they buy, for example, a new phone. Let’s assume you and I were to buy the latest iPhone, but you find a bargain and get it for 50% of the price that I am paying. We end up with the same phone, but an economist looking only at the differences in our spending would conclude that I am twice as well off since I am spending twice as much. Clearly, this is misleading.
Price differences are particularly important if households with less money to spend pay systematically lower prices than their high-spending counterparts. Previous studies have shown that the less well-off indeed find at least two ways to reduce their prices. First, similar to the example given above, people can shop for bargains and exploit the fact that prices for identical products can differ across stores (Aguiar and Hurst, 2007; Kaplan and Menzio, 2015). Second, people can substitute among very similar alternatives and buy, for example, the store-brand of milk rather than a branded option (Handbury, 2021). Less well-off households tend to pay lower prices for the same items and buy cheaper alternatives.
How much do these price differences matter for differences in households’ spending? To answer this question in a recent research paper, I use data from the NielsenIQ Consumer Panel, capturing price and quantity information for US households’ grocery purchases. To be able to compare prices across products, I sort them into groups of similar alternatives (for example, fresh apples being one group, and fresh oranges a second). With this sorting in hand, I measure how much the two margins outlined above – price differences for the same variety of a good and price differences across similar varieties – contribute to explaining inequality in expenditures across households.
The results are displayed below (Figure 1). Relative to the average price for a good, the lowest quintile of the expenditure distribution reduces total spending by 1.5% by paying less for identical products (within variety) and an additional 4% by buying less expensive products among close substitutes (across varieties). These savings decline as we go up the expenditure distribution, with the top quintile paying above-average prices for identical items and buying substitutes that are above-average expensive. Overall, price differences among similar items account for about 9% of the differences in spending between the top and bottom quintile, 2 percentage points of which are due to price differences for identical items.
Figure 1: Price differences by expenditure quintile (data)
Source: Nord (2022)
As striking as these price differences are, we cannot immediately translate them into people’s living standards. While it is clearly better to pay less for the same item, less well-off households have to make an effort to find bargains, and this takes time. So, while paying a lower price for their shopping basket, people end up with less free time for leisure at their disposal – they pay with time instead of money.
When it comes to differences in prices across varieties, interpretation is even harder. Households pay a different price but now also obtain a different product which they might value differently. To interpret these price differences across products, it is important to understand why a good is more expensive – whether the higher price reflects higher production costs of a higher-quality variety, or whether it is driven by retailers’ ability to discriminate in their price setting between products sold to households at different points of the income/wealth distribution.
These issues are best addressed within a theoretical framework, allowing a structured analysis of the interactions between households’ and retailers’ choices while matching key features of the data. Such a theory predicts that retailers respond to who is buying the product they are selling and set their prices according to the average buyer of a given good. Retailers perceive differences in households’ shopping behaviour as differences in the level of competition. If households shop more intensely, they are more likely to compare prices across stores. With more households comparing prices, it becomes more important for retailers to make a cheap offer relative to their competitors, bidding down overall prices for the product. As less well-off households (who shop around more intensely) buy a different basket of products relative to their better-off counterparts, this force is stronger for the basket of the poor and amplifies the direct effect of shopping – the less well-off paying less for any given good.
Calibrating the theory to the empirical results provides an estimate for how much this effect contributes to price differences across households. Within the theoretical framework, all differences in prices for identical products can be attributed to the direct effect of shopping – paying more or less for the same product. In addition, some of the price differences across products are due to the indirect effect that differences in shopping have on retailers’ pricing strategies. All remaining price differences are attributed to differences in production cost, referred to below as the ‘cost of quality’.
Figure 2 provides the results of this exercise. The left panel reproduces Figure 1 in the calibrated model and shows that it fits the data well when it comes to price differences within and across products. The right panel combines the effect of households’ shopping behavior on prices for identical items and across products and compares them to the cost of quality. Overall, about half of the price differences for similar items can be attributed to the direct and indirect effect of households’ shopping behaviour – accounting for 4.5% of inequality in spending between the lowest and highest quintile of expenditures. Through the theory, we can interpret these price differences as a 4.5% reduction in inequality in living standards relative to an economy without such differences in shopping behaviour. While less well-off households shop more intensely for cheap prices, better-off households buy more items and therefore must shop more often. Both require effort and lead to the similar losses in leisure along the distribution.
Figure 2: Price differences by expenditure quintile (theory)
Source: Nord (2022)
The same mechanism that amplifies the dampening effect of shopping on inequality also has implications for the response of the economy to fluctuations in aggregate income. As much as retailers respond to who the average buyer is for different goods, they also respond to how the average buyer in an economy changes over time. As some households are more exposed to, for example, an increase in unemployment or a decline in house prices, they have to reduce their consumption relative to those less affected – and now account for a lower share in demand. How much they shop becomes less relevant for retailers who now target their prices to a different average buyer. These shifts in demand composition can reconcile previous empirical estimates, which indicate a reduction in retail prices following a decline in local house prices, but little response of retail prices to local unemployment rates (Stroebel and Vavra, 2019; Coibion et al., 2015).
What does the link between inequality and prices imply for policy? First, policy-makers should be careful when interpreting numbers on inequality, paying close attention to price differences across households and their implications for the living standard these households can afford. Second, it is important to note that policy itself shapes the distribution of income and wealth and in this way influences prices in the economy. Just like the aggregate changes in income discussed above, policy can redistribute income across households and change who the average buyer in the economy is. By supporting incomes at the bottom of the distribution, government interventions can increase the average shopping effort in the economy and lower prices for all consumers as retailers respond to the increased competition.
Understanding how consumers in different income groups choose how to spend is vital for understanding how best to design policies to support those most in need. The importance of prices, and where they come from, cannot be understated.
Author: Lukas Nord
Author’s Note: Researcher(s)’ own analyses calculated (or derived) based in part on data from Nielsen Consumer LLC and marketing databases provided through the NielsenIQ Datasets at the Kilts Center for Marketing Data Center at The University of Chicago Booth School of Business. The conclusions drawn from the NielsenIQ data are those of the researcher(s) and do not reflect the views of NielsenIQ. NielsenIQ is not responsible for, had no role in, and was not involved in analyzing and preparing the results reported herein.