From commentary by Carol A. Stone of Nomura Securities International
BridgeNews NEW YORK--"A recession is a significant decline in activity spread across the economy, lasting more than a few months, visible in industrial production, employment, real income and trade." Robert Hall, Director of Studies in Economic Fluctuations at the National Bureau of Economic Research, gave us this definition of recession in a recent memo on the NBER's Web site.
Beginning and ending dates of recessions and expansions are delineated not by any government agency but by the NBER, a private non-profit organization of academic economists. Logically enough, it is the group's "Business Cycle Dating Committee" which determines these key dates. The committee has no regular meeting schedule; it gathers only when its members believe there is good reason to deliberate on the state of the economy.
The objective of Hall's memo is to explain why the Business Cycle Dating Committee had not yet met, despite the seeming intensity of the current slowdown. This definition of recession, noticeably, is not expressed in terms of quarters of declining GDP.
The common rule-of-thumb is that "a recession is two quarters of falling GDP." All recessions in the post-World War II era have contained two such quarters, not necessarily consecutive, but since GDP is derived from other more fundamental data, its movements are a consequence of business cycle developments, not a means of their definition.
Instead, Hall explains, the Bureau's Committee examines monthly data on the basic processes in the economy, and when they trace a significant and sustained decline, then the Committee declares a turning point of the economy into recession. Production, employment, income and trade (in the broad sense of distribution, not just foreign trade) are all major contributors. Measures of these items comprise "the coincident index," compiled by another economic research group, the Conference Board.
Turns in the coincident index largely demarcate phases of business cycles. Hall further explains that sustained weakness is part of the essence of recession, so the Committee always waits to name a turning point until six months of decline are evident. This past week, two of these basic indicators were reported, and all of them have been weak recently. Industrial production for May dropped 0.8 percent and was revised downward in the two prior months, now to show an especially sharp 0.6 percent fall in April.
Moreover, those two latest declines are widespread among industries. In April, the associated diffusion index was only 38.9, showing that production was rising at less than 40 percent of industries.
May's diffusion index is not yet tabulated, but aggregate data suggest it will be even worse. Production has fallen for eight straight months by a total of 4 percent. By comparison, in the 1990-91 recession production fell six months for a total of 4.6 percent. While we and others have made much of this weakness in industrial production, it is not enough for the NBER to declare a recession. Hall a even argues that it "measures the output of a small and declining part of the economy."
So he needs more to convince him that weakness pervades a broad swath of the economy. Business sales for April fell 0.5 percent in current dollars, their fourth consecutive drop. Although not yet available in inflation-adjusted terms for April, that month is expected to be weaker still.
However, the April fall occurred only at manufacturers, while sales rose 0.3 percent at wholesalers and 1.5 percent at retailers, an upward revision. Thus persistent, pervasive downtrends have not yet been established. Further, Hall asserts that retail sales are actually the most important component. In real terms, they had fallen in February and March, but showed a resurgence for April of 0.6 percent to a new peak.
Nominal retail sales were up only 0.1 percent in May, so in the face of a 0.4 percent CPI, their real value no doubt fell anew. But the most this seems to say is that the retail sector is hesitating and somewhat erratic, not that it is clearly tracing some longer-lasting decline. This pattern is also evident in employment, the other major input to a cycle-dating decision. Payrolls fell markedly in April, but only marginally in May, causing the NBER official to discount the employment weakness as an indicator of "recession."
Lastly, real disposable income is also an important input in describing the state of the economy. So far, it has seen only scattered decreases, although the softness in hours worked suggests it may well have dipped in May.
We recently discussed the current downtrend in hours worked. By May, they had fallen 0.5 percent from their peak last January. We pointed out that declines of such magnitude have generally occurred only during recession periods.
However it is not clear from the behavior of other gauges of economic activity, especially the sales, employment and income measures the NBER says are important, that recessionary conditions permeate the economy. All this said, public policy has already been applied as if the economy needed a boost. Policy-makers acted early, trying to forestall a full-blown contraction. The recessionary process may not be over. In addition to the basic data this week on sales and production, the Federal Reserve's Beige Book described the economy as slowing in most districts, with labor markets easing everywhere.
Not coincidentally, unemployment insurance claims remained in their new higher range. Weekly chain store surveys reported that sales got off to a weak start in June. Earnings reports from many companies continued to talk down profits prospects. Stock prices fell. Thus while the NBER's Hall has outlined why the Business Cycle Dating Committee has not yet met, the point is that they haven't. We may just be overly impatient for such a meeting to occur.
It will be a race to see if mounting pressures pushing the economy toward recession can be stayed by the positive forces of stimulative monetary and fiscal policy. End CAROL A. STONE is deputy chief economist of Nomura Securities International Inc., New York. Her views are not necessarily those of Bridge News, whose ventures include the Internet site www.bridge.com .