Unfortunately, this survey suffers from a bad shortcoming: it never bothered to ask if firms set prices on the basis of average unit costs plus a profit mark-up.
Nevertheless, it does report evidence that supports other data from surveys on mark-up pricing.
Apel et al. (2005: 331) find that the “typical” Swedish firm is operating in an oligopolistic market.
When asked to report how often prices were changed, the weighted results were that 40.3% of firms change prices once per year, and 27.1% adjust their prices less than once a year (Apel et al. 2005: 318).
When given a list of reasons for price rigidity, the New Keynesian theories of “menu costs” and “information-gathering costs” ranked as “totally unimportant” and “of minor importance” (Apel et al. 2005: 329).
The survey asked the following question:
“We … asked: ‘Assume that you notice that there has been a slight increase in demand for your main article/service. What is normally the strongest argument for leaving the price unchanged?’ The respondents were given the following alternatives: (1) it is too costly to change the price (relabeling, new price lists, etc.), (2) it is important not to diverge from the prices of competitors, and (3) it is better to leave the price unchanged as long as the costs do not change. An overwhelming majority chose one of the latter two alternatives, with more or less equal shares given to each of these explanations. In fact, the turnover-weighted estimate of the proportion that considered actual costs of changing prices to be the most important factor was 0.2%.” (Apel et al. 2005: 323).This strongly suggests that cost-based/mark-up pricing was a widespread price setting behaviour that led to these results.
Also consistent with other surveys is that implicit and explicit contracts are important sources of price rigidity (Apel et al. 2005: 324, 327).
In trying to determine whether mark-ups over costs are procyclical or countercyclical, some interesting evidence was found:
“the respondents were asked to rank how well a number of statements described the development of markups over the business cycle. Let us first note that marginal cost is difficult to estimate, except for very simple production technologies, and that firms’ pricing decisions often tend to be based on average variable costs. For this reason, we are unwilling to draw any strong conclusions about how markups over marginal costs develop over the business cycle based on the answers to this question. An increase in marginal costs is clearly associated with an increase in average variable costs, but the relationship between marginal and average variable costs is not necessarily one-to-one. For instance, if we increase quantity so that we move from a relatively flat section of the marginal cost curve to a steeply upward-sloping section, marginal costs will rise sharply whereas average variable costs will only gradually reflect the higher marginal costs.Now the finding that marginal cost is difficult for many businesses to even calculate and that firms face serious “difficulty estimating the marginal cost for all but the simplest techniques” (Apel et al. 2005: 330, n. 20) is consistent with other studies. So too is the finding that price changes in mark-up prices are mainly cost-driven.
As reported in Table 5, the most common practice seems to be the use of a constant markup, changing the price proportionally when costs change.” (Apel et al. 2005: 330).
But Apel et al.’s implied conclusion that mark-up prices are simply based on “average variable/direct costs” is untrue, and shows how many neoclassical economists who design these surveys do not even understand mark-up pricing conventions. Most mark-up prices are ultimately based on total average unit costs (including both average variable and fixed/overhead costs), because even when mark-up prices are initially based on average variable costs, the crucial point is that the mark-up will include both average unit fixed/overhead costs and an allowance for profit.
That is, while technically in initial “costing,” firms begin with average variable/direct costs, they add average unit fixed/overhead costs to this (Lee 1998: 10, 204–205), so that the fundamental cost concept is total average unit costs.
The failure to understand these facts causes deep confusion, and the utterly unsound idea that the firms simply use average variable/direct unit costs, and that this is a good general proxy for marginal cost. Both ideas are wrong – and this cannot be stressed enough.
Apel, Mikael, Friberg, Richard and Kerstin Hallsten. 2005. “Microfoundations of Macroeconomic Price Adjustment: Survey Evidence from Swedish Firms,” Journal of Money, Credit and Banking 37.2: 313–338.
Lee, Frederic S. 1998. Post Keynesian Price Theory. Cambridge University Press, Cambridge and New York.