Critics assert that efforts to expand
the scope of the public sector will drive up interest
rates and crowd out private business investment.
The accusation is particularly likely to be heard
when a proposal explicitly foresees the use of the
credit market, deficits, and public debt to finance
the expansion.
Are these fears justified? There is a two-part
answer to this question, the first related to economic
theory, and the second to the specific conditions facing
the United States in the world credit markets.
The theory of “crowding out” is based on a
common misconception of the nature of savings in
our economy, namely the idea that savings are a
“pool,” fixed in size, from which the public and
private sectors alike draw to finance their desired
The Macroeconomic Considerations of a Public Investment Strategy
rates of spending. No such pool exists. Rather,
what we measure as savings is created after the fact,
by the spending decisions of governments and private
businesses. These decisions create income; the
difference between income and consumption (the
latter, strongly established by habit), is savings.
Historically, savings have tended to rise in good
economic times and fall in bad economic times
because the household sector had limited direct
access to the credit markets. Thus a public budget
deficit was practically required for an expansionary
policy, and this gave rise to the idea that the public
sector competes with the private sector for available
funds. In the 1990s, however, the combination of a
stock market bubble, a housing boom, and credit
market innovations lifted this constraint, especially
for upper-income households. The household sector
was able to draw, to an unprecedented extent,
on loans against home equity (or in some cases,
against capital gains in the stock market) to finance
current consumption spending. Personal savings
went negative as the economy boomed.
A fall in personal savings was thus the counterpart
of the consumption boom. The fall in savings
also sent the federal budget into surplus: as spending
rose, incomes rose, and tax revenues grew. This
was, in effect, a “Keynesian devolution”: the economy
moved toward full employment on the
strength of spending fostered by public policy and,
in some important respects, guaranteed by the
public sector, but not carried out directly by government.
Thus the private (household and corporate)
sector ran deficits that the government would,
under other conditions, have been obliged to run.
Were the interest rate determined mainly by the
pull of federal credit demands on a “pool of savings,”
interest rates should have fallen as the federal
budget went into surplus. That did not happen.
Long-term interest rates remained around 7
percent throughout the period, and only declined
toward present values after the Internet bubble
burst in 2000. They reflected, in other words, the
health of the economy as a whole, and the use of
resources in relation to full capacity. After the
NASDAQ collapse, and especially following 9/11,
long-term interest rates fell sharply, despite the
strong movement of the federal budget into deficit.
If a public investment program (in combination
with other policies) were so successful as to bring
the economy up to full employment in the near
term, it might have a modest effect on long-term
interest rates. However, during our last experience
with noninflationary full employment in the late
1990s, there were few complaints that long-term
interest rates, which were then 200 basis points
higher than they are now, were a constraint on private
business activity. There is no strong reason to
believe that in a new period of full employment
this situation would be different. Moreover, the
scenario of a return to full
employment is optimistic.
It is now clear that the
depressing effects of the
slowdown in housing and
the fallout from the subprime
mortgage crisis will
be felt in a slower growth
rate of real GDP—perhaps
even in a recession. In
these circumstances, a
major public investment
strategy is exactly what will
be needed to stimulate job creation and private
investment. However, there is no reason whatsoever
to fear that even a large public investment
program would raise interest rates significantly.
We can conclude, first, that there is no direct
connection between federal budget deficits or surpluses
and long-term interest rates. Bond-financed
public investment poses no significant threat to
financial stability on that account. We can also
conclude that long-term interest rates may be
influenced by a combination of capacity pressures
in the domestic economy and inflation expectations.
However, this effect would be limited by the
fact that the domestic economy is unlikely to be at
full capacity in any event, and by conditions in the
world economy, which determine total effective
supply as well as the price for commodities and the
value of the dollar.