Cost Pass-Through in the U.S. Coffee Industry
by Ephraim Leibtag
, Alice Nakamura, Emi Nakamura, and Dawit Zerom
Economic Research Report No. (ERR-38) 28 pp, March 2007
A perennial issue in economics is the effect of changes in
commodity prices on manufacturer and retail food prices. The
traditional explanation is that the extent to which cost increases
are "passed through" in a vertically organized production process
depends on the market power of producers at each stage of
production as well as the value added by each producer in the
production process. The U.S. coffee industry is an excellent venue
to study the issue of cost pass-through, since green coffee beans
are important components of the marginal costs in this industry and
are publicly traded commodities.
What Is the Issue?
This report uses unique data from the U.S. coffee industry to
estimate how changes in commodity costs affect retail coffee
prices. The results are relevant beyond the coffee industry,
providing insight into how changes in commodity costs pass through
to consumer and producer prices in other industries, too. "Cost
pass-through" is a central issue in international economics since
it determines how an economy responds to exchange rate adjustments
as well as to changes in the prices of other imported commodities,
such as oil.
What Did the Study Find?
Average manufacturer coffee prices dropped from 23 cents in 1997
to 17 cents per ounce in 2002. That drop corresponded with a fall
in the coffee-bean share of the manufacturer price from 48 percent
to 24 percent, while labor and other material costs rose from 15
percent to 32 percent.
The authors found that, on average, a 10-cent increase in
green-coffee-bean prices per pound yields a 2-cent increase in both
manufacturer and retail prices in the current quarter. If a cost
change persists, it will be incorporated into manufacturer and
retail prices approximately cent-for-cent with the commodity cost
change. In addition, cross-sectional differences in prices are
substantially larger at the retail than the wholesale level.
Since manufacturer prices adjust approximately one-for-one with
commodity prices (rather than proportionally), the ratio between
manufacturer prices and commodity costs rises as commodity costs
rise. The study does not find robust evidence that coffee prices
respond more to increases than to decreases in costs.
How Was the Study Conducted?
An unusually rich collection of data on the ground-coffee
industry was used to analyze the issue of cost pass-through. The
data set included market-level average retail prices collected by
Nielsen ScanTrack, market- level manufacturer prices collected by
Promodata, and panel data collected by Nielsen Homescan to
calculate the share of coffee by brand for each income level.
Regression analysis was used to estimate the impact of changes in
commodity prices on retail and manufacturer prices. These
regressions are carried out for both absolute levels and in
percentage terms. In addition, instrumental variable and
fixed-effect methods were used to look at the manufacturer-retail
price relationship and to analyze whether prices respond
asymmetrically to cost increases and decreases.