ESSAYS ON ISSUES
THE FEDERAL RESERVE BANK
DECEMBER 1994
NUMBER 88
OF CHICAGO
Chicago Fed Letter
The ups and downs of
commodity price indexes
The current concern about inflation
began with the run-up in commodity
prices at the beginning of 1993. At
that time, financial markets overreact-
ed when they interpreted a temporary
surge in commodity price indexes as
a sign of imminent higher inflation.
As it turned out, commodity prices
were responding to a variety of short-
lived economic events and, contrary
to expectation, inflation actually de-
clined in 1993.
More recently, the robust growth in
domestic economic activity since late
1993 has caused some pressure on the
prices of some industrial materials.
As a result, the spotlight is once again
on commodity price indexes as lead-
ing indicators of inflation. Commodi-
ty-based indicators are calculated as
an average of the prices of different
commodities, and potentially trans-
late individual price movements into
a common measure of aggregate
price changes.
Spot and futures prices of individual
commodities are determined and
quoted daily in competitive auction
markets; these prices adjust quickly to
changes in supply and demand. Com-
modities account for only a small
fraction of the cost of finished goods.
Yet because they have a considerable
weight in Consumer Price Index
(CPI) calculations, a continued in-
crease in commodity prices may push
up the inflation rate, as measured by
the percent change in the CPI. Thus
changes in materials prices can be
real-time indicators of other price
changes, current or anticipated.
A considerable amount of time may
pass, however, before commodity
price gains translate into higher infla-
tion. Furthermore, price increases in
industrial commodities and raw mate-
rials don’t always cause inflation to
rise. Sometimes they are only tempo-
rary responses to a variety of events
whose effects reach no further. Also,
since commodity price indexes re-
spond to changes in supply and de-
mand of individual commodities, they
may reflect price fluctuations that are
only relative and not indicative of
inflationary pressures.
The November 1993 Chicago Fed Letter
showed that inflation forecasts based
on individual commodity prices and
commodity price indexes can be high-
ly misleading, since commodity prices
often signal concurrent changes in
price and output.1 In this Fed Letteryfe
take the analysis a step further and
present evidence that commodity
price indexes are not statistically use-
ful in predicting consumer price infla-
tion. First, we analyze the composi-
tional characteristics of three
different commodity price indexes
designed specifically to help forecast
inflation. Then we present the results
of a number of statistical tests we
performed to assess the indexes’ pow-
er to do just that. The
tests indicate that as
forecasters of inflation,
commodity price index-
es contribute no addi-
tional information be-
yond what is contained
in the past history of
consumer prices.
How are the indexes
composed?
We analyzed the three
most widely known
commodity price index-
es: the Commodity
Research Bureau Fu-
tures Price Index
(CRB), the Journal of
Commerce Industrial Price Index
QOCCI), and the Change in Sensitive
Materials Prices (SMPS) .2 Their main
distinguishing characteristics are the
commodity price used (futures or spot
prices), the number of component
commodities, and the weight attached
to each commodity to calculate the
index. As figure 1 shows, CRB is cal-
culated on the basis of futures prices
of 21 commodities, JOCCI is calculat-
ed on spot prices of 18 industrial
commodities, and SMPS is calculated
on spot prices of 12 crude and inter-
mediate materials and 13 raw industri-
al materials. Furthermore, CRB and
SMPS assign equal weights to their
components, while JOCCI assigns
individual weights based on the com-
ponents’ estimated ability to lead
consumer price inflation.
One major shortcoming of these com-
modity price indexes is the weighting
scheme used to calculate them.
When commodities are equally
weighted, as they are in CRB and
SMPS, for example, a 1% increase in
the price of cocoa would have the
same impact on the index as a 1%
1. Composition of commodity price indexes
|
CRB |
JOCCI |
SMPS | |
|
Prices |
futures |
spot |
spot |
|
Components |
21 |
18 |
25 |
|
Weights |
equal |
individual |
equal |
|
Weights by | |||
|
category | |||
|
Metals |
19% |
35% |
38% |
|
Energy |
14% |
12% |
0% |
|
Livestock |
14% |
0% |
0% |
|
Grains, food, | |||
|
and fiber |
43% |
17% |
29% |
|
Other |
10%a |
36%b |
33%c |
aOrange juice and lumber
bRubber, red oak, hides, tallow, boxes, and plywood
cRubber, hides, rosin, tallow, wastepaper, sand, and lumber