In the period immediately prior to September 11, there were tentative
signs that some sectors of the U.S. economy had begun to stabilize, contributing
to a hope that the worst of the previous cumulative weakness in world
economic activity was nearing an end. That hope was decisively dashed
by the tragic events of early September. Adding to the intense forces
weighing on asset prices and economic activity before September 11 were
new sources of uncertainty and risk that began to press down on global
demand for goods and services.
In almost all areas of the world, economies weakened further, a cause
for increasing uneasiness. The synchronous slowing in activity raised
concerns that a self-reinforcing cycle of contraction, fed by perceptions
of greater economic risk, could develop. Such an event, though rare, would
not be unprecedented in business-cycle history.
We had already observed a coincident deceleration in activity among the
world economies over the past year, owing apparently, at least in part,
to the retrenchment in the high-technology sector. The global nature of
most technology industries and the global reach of the capital markets
in which the firms in these industries are valued and funded appears to
have fostered a greater synchronousness in world activity in this cycle,
seemingly broader than has generally been the case. However, before the
terrorist attacks, it was far from obvious that this concurrent weakness
was becoming self-reinforcing.
But, if ever a situation existed in which the fabric of business and consumer
confidence, both here and abroad, was vulnerable to being breached, the
shock of September 11 was surely it. Indeed, for a short period, in response
to that shock, U.S. economic activity did drop dramatically.
But, arguably, our economy has not been weakening cumulatively in recent
weeks. In fact, indications of stabilization, similar in many respects
to those observed in the period immediately preceding September 11, have
been appearing with greater frequency. A possible significant contributor
to this emergence of stability--if that is what it is--may be the very
technologies that have fostered coincident global weakness: those that
have substantially improved access of business decisionmakers to real-time
Thirty years ago, the timeliness of available information varied across
companies and industries, often resulting in differences in the speed
and magnitude of their responses to changing business conditions. In contrast
to the situation that prevails today, businesses did not have real-time
data systems that enabled decisionmakers in different enterprises to work
from essentially the same set of information. In those earlier years,
imbalances were inadvertently allowed to build to such an extent that
their inevitable correction engendered significant economic stress. That
process of correction and the accompanying economic and financial disruptions
too often led to deep and prolonged recessions.
Today, businesses have large quantities of data available virtually in
real time. As a consequence, they address and resolve economic imbalances
more rapidly than in the past. At the same time, firms are largely operating
with the same information set, and thus resolution of imbalances induces
parallel movements in activity. Contractions initially may be steeper,
but because imbalances are more readily contained, cyclical episodes overall
should be less severe than would be the case otherwise.
In the current situation, inventories, especially among producers and
purchasers of high-tech products, did run to excess over the past year,
as sales forecasts went badly astray; alas, technology has not allowed
us to see into the future any more clearly than we could previously. But,
technology did facilitate the quick recognition of the weakening in sales
and backup of inventories. This enabled producers to respond forcefully,
as evidenced by output adjustments that have resulted in the extraordinary
rate of inventory liquidation currently under way.
Inventories in many industries have been drawn down to levels at which
firms will soon need to taper off their rate of liquidation, if they have
not already done so. Indeed, in recent months, there have been fewer reports
from industrial purchasing managers that their customers' inventories
are too high. Moreover, the relative stability of industrial commodity
prices in recent weeks, and especially the recent firmness in the prices
of semiconductors, could be hinting at less intense stock drawdowns.
A slowing in the rate of inventory liquidation will induce a rise in industrial
production if demand for those products is stable or is falling only moderately.
That rise in production will, other things being equal, increase household
income and spending. The runoff of inventories, even apart from the large
reduction in motor vehicle stocks, remained sizable in the fourth quarter.
Hence, with production running well below sales, the potential positive
effect of the inevitable cessation of inventory liquidation on income
and spending could be significant.
But that impetus to activity will be short lived unless the demand for
goods and services itself starts to rise. On that score, despite a number
of encouraging signs of stabilization, it is still premature to conclude
that the forces restraining economic activity here and abroad have abated
enough to allow a steady recovery to take hold. For that to happen, sustained
growth of final demand must kick in before the positive effects of the
swing from inventory liquidation to accumulation dissipate.
For the household sector, which had been a major stabilizing force through
most of last year's slowdown, the outlook for demand is mixed. Low mortgage
interest rates and favorable weather have provided considerable support
to homebuilding in recent months. Moreover, attractive mortgage rates
have bolstered both the sales of existing homes and the realized capital
gains that those sales engender. They have also spurred refinancing of
existing homes and the associated liquification of increases in house
values. These gains have been important to the ongoing extraction of home
equity for consumption and home modernization.
The recent rise in home mortgage rates, however, is likely to damp housing
activity and equity extraction. It is already having an effect on cash-outs
from refinancing. Cash-outs rose from an estimated annual rate of about
$20 billion in early 2000 to a rate of roughly $75 billion in the third
quarter of last year. But the pace of cash-outs has likely dropped noticeably
in response to the recent decline in refinancing activity that has followed
the backup in mortgage rates since early November.
Consumer spending received a considerable spur from the sales of new motor
vehicles, which were remarkably strong in October and November owing to
major financing incentives. Sales dropped last month when the incentives
were scaled back, but have remained surprisingly resilient. Other consumer
spending appears to have advanced in recent months, though at a subdued
The substantial declines in the prices of natural gas, fuel oil, and gasoline
have clearly provided some support to real disposable income and spending.
These price declines added more than $50 billion at an annual rate to
household purchasing power in the second half of last year. However, a
decline in energy prices provides, in effect, only a one-shot boost to
consumption, albeit one that is likely to take place over time. To have
a more persistent effect on the ongoing growth of total personal consumption
expenditures, energy prices would need to continue to decline. Futures
prices do not suggest that such a decline is in the immediate offing,
but the forecast record of these markets is less than sterling.
Although the quantitative magnitude and precise timing of the wealth effect
remain uncertain, the steep decline in stock prices since March 2000 has,
no doubt, curbed the growth of household spending. Although stock prices
recently have retraced a portion of their earlier losses, the restraining
effects from the net decline in equity values presumably have not, as
yet, fully played out. Future wealth effects will depend importantly on
whether corporate earnings improve to the extent currently embedded in
Perhaps most central to the outlook for consumer spending will be developments
in the labor market. The pace of layoffs quickened last fall, especially
after September 11, and the unemployment rate rose sharply. Over the past
month or so, however, initial claims for unemployment insurance have declined
markedly, on balance, suggesting some abatement in the rate of job loss.
Although this development would be welcome, the unemployment rate may
well continue to rise for a time, and job losses can be expected to put
something of a damper on consumer spending. However, the extent of that
restraint will depend on how much of any rise in unemployment is the result
of weakened demand and how much reflects strengthened productivity. In
the latter case, average real incomes could rise, at least partially offsetting
losses of purchasing power that stem from diminished levels of employment.
Finally, economic policies will have an important influence on household
spending in the period ahead. No doubt, we will continue to benefit from
the tendency of our tax and entitlement systems to buffer cyclical swings
in income. Moreover, despite the failure of Congress to enact further
tax cuts and spending increases, the continued phase-in of earlier reductions
in taxes and the significant expansion of discretionary spending already
enacted should provide noticeable short-term stimulus to demand.
Some of this stimulus has likely been offset by increases in long-term
market interest rates, including those on home mortgages. The recent rise
in these rates largely reflects the perception of improved prospects for
the U.S. economy. But over the past year, some of the firmness of long-term
interest rates probably is the consequence of the fall of projected budget
surpluses and the implied less-rapid paydowns of Treasury debt.
In our conduct of monetary policy, the Federal Reserve responded to the
weakening economy over the past year by markedly lowering our target for
the federal funds rate. We accelerated the pace of rate reductions during
this period in response to the accelerated pace of economic adjustment.
Moreover, the magnitude of policy adjustment and the resulting low level
of the federal funds rate responded both to the strength of the forces
restraining demand and to the continued subdued pace of underlying inflation.
Liquidity, as a consequence, has expanded significantly, and the accompanying
lower interest rates have supported spending and held down the cost to
households of servicing debts.
The dynamics of inventory investment and the balance of factors influencing
consumer demand will have important consequences for the economic outlook
in coming months. But, the broad contours of the present cycle have been,
and will continue to be, driven by the evolution of corporate profits
and capital investment.
The retrenchment in capital spending over the past year was central to
the sharp slowing we experienced in overall activity. The steep rise in
high-tech spending that occurred in the early post-Y2K months was clearly
not sustainable. The demand for many of the newer technologies was growing
rapidly, but capacity was expanding even faster, exerting severe pressure
on prices and profits. New orders for equipment and software hesitated
in the middle of 2000, and then fell sharply as firms re-evaluated their
capital investment programs. Uncertainty about economic prospects boosted
risk premiums significantly, and this rise, in turn, propelled required,
or hurdle, rates of return to markedly elevated levels. In most cases,
businesses required that new investments pay off much more rapidly than
they had previously. For much of last year, the resulting decline in investment
outlays was fierce and unrelenting. Although the weakness was most pronounced
in the technology area, the reductions in capital outlays were broad-based.
These cutbacks in capital spending interacted with, and were reinforced
by, falling profits and equity prices. Indeed, a striking feature of the
current cyclical episode relative to many earlier ones has been the virtual
absence of pricing power across much of American business, as increasing
globalization and deregulation enhanced competition. In this low-inflation
environment, firms have perceived very little capability to pass cost
increases on to customers. Growth in hourly labor compensation has slowed
in response to deteriorating economic conditions, but even those smaller
increases have continued to outstrip gains in output per hour for the
corporate sector on a consolidated basis. The result has been that profit
margins are still under pressure.
Business managers, with little opportunity to raise prices, have moved
aggressively to stabilize cash flows by trimming workforces. These efforts
have limited the rise in unit costs, attenuated the pressure on profit
margins, and ultimately helped to preserve the vast majority of private
sector jobs. To the extent that businesses are successful in stabilizing
and eventually boosting profits and cash flow, capital spending should
begin to recover more noticeably.
Such success would likely be accompanied by a decline in elevated risk
premiums back to more normal levels and, with real rates of return on
high-tech equipment still attractive, should provide an additional spur
to new investment. When capital spending eventually recovers, its growth
is likely to be less frenetic than that which characterized 1999 and early
2000, when outlays were boosted by the dislocations of Y2K and the extraordinarily
low cost of capital faced by many firms.
Still, the evidence strongly suggests that new technologies will present
ample opportunities to earn enhanced rates of return. Indeed, anecdotal
reports from businesses around the country suggest that the exploitation
of available networking and other information technologies was only partially
completed when the cyclical retrenchment of the past year began. Many
business managers are still of the view, according to a recent survey
of purchasing managers, that less than half of currently available new,
and presumably profitable, supply chain technologies have been put into
While these opportunities remain abundant, they will now play out against
the backdrop of a major uncertainty that we all must deal with these days--the
specter of further terrorist incidents on American soil. It simply is
not possible to predict whether there will be any such incidents or to
forecast their possible consequences for the economy. But we can have
little doubt that the tragic events of September 11 have left obvious
marks on the economy that will not soon fade even though some of the initial
impact of the shock has receded. Importantly, as I suggested shortly after
the event, adjustments to new levels of perceived risk will cause a one-time
downward shift in the level of productivity.
Clearly, businesses will be less comfortable now than they were before
September 11 in allowing inventories to shrink to minimal levels in a
just-in-time supply chain. Moreover, in some industries, resources will
need to be diverted from efficiency-enhancing capital investment to providing
security and contingency backup. Fragmentary data for the months following
September 11, however, indicate output per hour is holding up well. Temporary
labor-shedding may have overwhelmed the effects of added security and
redundancy. It is not yet clear whether the negative shock to output per
hour from the heightened risks is small, or just delayed. In any event,
once these adjustments are completed, the full benefits of more rapid
technological advance should show through to the growth of productivity.
The central role that is being played by technological advance poses special
challenges to forecasters. Few technologies that influence our economic
future are truly anticipated much in advance. And even when they are anticipated,
their effect on economic growth is difficult to predict, in part because
their pace of diffusion and application is so uncertain. The latter consideration
is particularly significant to the longer-term rate of growth of productivity.
The events of the past decade clearly illustrate those difficulties. Few
observers foresaw how microprocessors, integrated circuits, and the mating
of laser technology with fiber optics, even well into their development
and application, would rejuvenate the American economy.
For example, as recently as a decade ago, the outlook was for a continuation
of meager gains in output per hour, with the rate of growth barely exceeding
one percent per year, if that. Instead, during the last half of the 1990s,
we experienced a surge in productivity growth well above the rate of increase
experienced in the previous quarter-century.
Even as our economy slipped into recession, the growth of output per hour
remained positive and, as I indicated earlier, has held up well even in
the wake of September 11. Until last year, the hypothesis of an accelerated
productivity trend had not been tested in the contracting phase of a business
cycle. Recent developments have provided that test, and the early returns
certainly look favorable to the hypothesis.
In retrospect, our economic structure changed in the mid-1990s. The crucial
agent of this remarkable change was the quantum leap in information availability.
If the tentative indications that the contraction phase of this business
cycle is drawing to a close are ultimately confirmed, we will have experienced
a relatively mild downturn. To be sure, a great deal of real economic
pain has been felt over the past year and a half. But imbalances have
not been allowed to fester. They could have progressively undermined endeavors
at stability and prolonged this difficult period.
The American economy has had to absorb some extraordinary shocks over
the past year and a half. For the economy to have weathered as well as
it has a severe deflation of equity asset values followed by an unprecedented
blow from terrorists to the foundations of our market systems is impressive.
In my judgment, this performance is a testament to the exceptional degree
of resilience and flexibility that our economy has gained in recent years,
much of which owes to advances not only in information technology, but
to the globalization and deregulation of our markets, as well. The adaptability
and resourcefulness of our businesses and workers have been especially
important in this trying period.
There are sound reasons for concluding that the long-run picture remains
bright, and even recent signals about the current course of the economy
have turned from unremittingly negative through the late fall of last
year to a far more mixed set of signals recently. But I would emphasize
that we continue to face significant risks in the near term. Profits and
investment remain weak and, as I noted, household spending is subject
to restraint from the backup in interest rates, possible increases in
unemployment, and from the effects of widespread equity asset price deflation
over the past two years.
But if the recent more favorable developments continue and gather momentum,
uncertainties will diminish, risk premiums will fall, and the pace of
capital investment increase. Should those gains in investment materialize,
they would, doubtless, embody the newest technologies. As we have witnessed
so clearly in recent years, advances in technology have enhanced the growth
of productivity, which, in turn, has been essential to lifting our standards