Economic Trends

Inflation And Prices

The Consumer Price Index rose 0.1% (1.3% annual rate) in July, the same rate of increase it posted in June. According to the Labor Department, the CPI indexes for food and energy, which were unchanged in June, rose 0.2% for food (2.1% annual rate) and 0.4% for energy (5.0% annual rate) in July. Moreover, the CPI communications index rose sharply during the month, partly because of a hike in postal rates.

Over the most recent 12 months, the CPI has risen at a modest 1.5% rate. Core measures of inflation, by contrast, continue to rise more rapidly. The CPI excluding food and energy, for instance, rose at a 2.3% rate over the past 12 months. During the same period, the trimmed-mean inflation measures--the median CPI and the 16% trimmed-mean CPI--rose 3.4% and 2.1%, respectively. Despite the faster rate of increase, however, all core measures have been trending down throughout the course of this year.

In July, the Labor Department issued a new index, the chained CPI, which is intended to improve on the conventional CPI by addressing the issue of substitution bias. Substitution bias arises when a price index fails to account for the way a change in (relative) prices might cause consumers to change how they allocate expenditures among the items in their market basket. The conventional CPI, for instance, assumes that the market basket remains fixed: No matter how much the price of one good may rise relative to others, consumers are assumed to continue buying these goods in the same relative quantities. The chained CPI, by contrast, uses a method that accounts for the fact that the consumers' market basket changes over time in response to changes in relative prices.

Estimates of the substitution bias inherent in the conventional CPI calculation have generally been lower than +0.5% annually. Indeed, a commission convened by the Senate Finance Committee in the mid-1990's concluded that inflation was being overestimated by 0.2%-0.4% annually. The chained CPI has been available only for the last few years, but simulations by the Bureau of Labor Statistics indicate that these bias estimates were appropriate for most of the 1990's. During the current decade, the gap has become much more pronounced, often nearly a full percentage point. In recent months, however, the difference between the year-over-year differential has narrowed to about 0.4%.

Chaining the core measure tells much the same story, with differences in annual growth between the chained and conventional indexes at or near a full percentage point in 2000 and 2001 and narrowing to 0.6% in the most recent several months. Interestingly, the measure that tracks the chained CPI most closely is the Personal Consumption Expenditures Price Index, which uses a similar chaining method.

Economists' consensus expectation of inflation is about 2.5% over the next 18 months. Households seem optimistic about the inflation outlook over the short run: Their year-ahead expectations of inflation have fallen for the third consecutive month. However, households are less sanguine about the inflation outlook for the next five years or so; these longer-run expectations have been trending up since the beginning of this year.

Monetary Policy

At its August 13 meeting, the Federal Open Market Committee left the target federal funds rate unchanged, although it altered the balance-of-risk statement "towards conditions that may generate economic weakness." The federal funds futures market now has built in a strong possibility of lower rates. With implied yields reaching a minimum of 1.59% in February 2003, the market seems quite confident of a 25 basis point cut by early next year.

The Taylor rule, one gauge of monetary policy, posits that the FOMC chooses the target rate as a balanced response to weakness and inflation. The form of the Taylor rule depends on the weights given to inflation and output, and to the assumed inflation target. Recently, the rule has correctly predicted the direction of changes in the federal funds rate.

Waiting can be the hardest part, but the eight and a half months since the target federal funds rate last moved is about the average period of no action over the past 20 years. The most recent rate reduction far exceeded the average, both in duration (11 months) and in the size of the decline (4.75%), although it set no records in either. Cumulative rate reductions have been larger and have taken longer to implement than cumulative rate increases. On average, however, periods when the FOMC has held rates steady have been the longest.

Money And The Financial Markets

The Federal Open Market Committee's August 13 statement indicated that the balance of risks for the economy tilted toward economic weakness, a change from its previous statement that economic weakness and inflation were evenly balanced. How do the financial markets view the current balance of risks? Put another way, do market participants see a 1.75% federal funds rate or an M2 growth rate of more than 5% as a sign of inflation?

Over the long term, the answer seems to be no. One market measure of expected inflation, the spread between yields on 10-year nominal Treasury bonds and 10-year Treasury inflation-indexed bonds, has fallen. In late May, the spread implied expected inflation exceeding 2%; it now implies values closer to 1.75%. In the short term, the answer again appears to be no. A measure of expected inflation over the next 30 days, derived from surveys and Treasury bill rates, suggests a rate of only 2.4%.

A less favorable indicator of inflation risk comes from the gold market, where prices have increased 21% since April 2001 and 12% since the beginning of this year. However, the price of gold is not an infallible sign of inflation because often it is affected by specific market factors such as central bank sales or jewelry demand.

A rise in gold prices often reflects a flight to security when the economic or political outlook becomes uncertain, but other measures of risk in the financial world do not point to uncertainty. The TED spread, the yield difference between Eurodollar deposits and Treasury bills, often picks up on such concerns because it measures credit risk at international banks without reflecting exchange rate risk; it remains very low. In the domestic market, the yield spread between 90-day commercial paper and three-month Treasury bills also remains quite low.

At the lower end of the credit spectrum, things look less rosy. Spreads over Treasuries of both high-yield and BBB-rated bonds have increased substantially in recent months. Thus, credit concerns seem to be growing, at least for lower-rated borrowers. Such borrowers become more important if we turn from rates to ratings. In any given year, some firms get stronger and others get weaker, but a good measure of the overall trend is the ratio between ratings upgrades (receiving a higher--that is, better--rating, which suggests the company has become less risky) and downgrades. Not only have downgrades outnumbered upgrades for the past several years, but the trend in the ratio has worsened as well.

A classic measure of risk in the economy is the term structure of interest rates coming out of the Treasury yield curve. The yield curve has moved little since last month, although it has steepened noticeably since this time last year, mainly because short rates have fallen. For most of 2002, however, short rates have held steady, with longer-term rates dropping 120 basis points since late spring.

In the past, a steep yield curve indicated robust economic growth. Plotting the 10-year, 3-month spread against GDP growth for the year ahead shows that the yield curve has been a fairly reliable signal since 1960, although periods of high growth occasionally are accompanied by a low spread. A negative spread (inverted yield curve) reliably indicates recessions, although, like many other signs, it was confused by the 1967 mini-recession. Thus, while the present steep yield curve may not guarantee a strong recovery, it suggests a low likelihood of a "double-dip" recession.