[Economic Report of the President (2010)]
[Administration of Barack H. Obama]
[Online through the Government Printing Office, www.gpo.gov]

 
CHAPTER 4

SAVING AND INVESTMENT


The United States appears poised to begin its recovery from the most
severe recession since the GreatDepression. But as discussed in
Chapter  2, the recession has been unusually deep, and the crisis has
caused declines in credit availability as well as weak consumer and
business confidence. As a result, achieving the private spending
necessary to support a robust and full recovery has been, and will
continue to be, challenging.

Moreover, as the President has repeatedly emphasized, it is not
enough simply to return to the path the economy was on before the
slump. The growth that preceded the recession saw high consumption
spending, low private saving, excessive housing construction,
unsustainable run-ups in asset prices (especially for assets related
directly or indirectly to housing), and high budget and trade
deficits. That path was unstable--as we have learned at enormous
cost--and undermined long-run prosperity. Thus, as the economy
recovers, a rebalancing will be necessary. The composition of
spending needs to be reoriented in a way that will put us on a path
to sustained, stable prosperity.
In thinking about the twin challenges of recovery and reorientation,
it is useful to consider the division of demand into its components.
Overall or aggregate demand can be classified into personal
consumption expenditures, residential investment, business
investment, net exports, and government purchases of goods and
services. Government purchases, which consist of such items as
Federal expenditures on national defense and state and local
spending on education,are relatively stable.
This is especially true when one recalls that government transfers,
such as spending on Medicare or Social Security, are not part of
government purchases but rather are elements of personal income.
Thus, it is the behavior of the remaining components that will be
central to addressing the challenges of generating enough demand for
recovery and a better composition of demand for long-run growth and
stability.
This chapter lays out a picture of how the components of private
demand behaved during the downturn and how they are likely to evolve
as the economy recovers and once it returns to full employment. The
chapter describes the transition that has already occurred away from
low personal savingand high residential investment, as well asthe
transition that needs to occur toward greater business investment and
net exports. It also describes the President's initiatives for
encouraging the transitions necessary for long-run prosperity and
stability.

The Path of Consumption Spending

Figure 4-1 shows the share of gross domestic product (GDP) that
takes the form of production of goods and services directly purchased
by consumers. The figure has two key messages. First, consumption
represents a substantial majority of output. As a result, movements in
consumption play a central role in macroeconomic outcomes. Second, the
fraction of output devoted to consumption has been rising over time,
leaving less room for components that contribute to future standards
of living. The behavior of consumption will therefore be central to
addressing both the shorter-run challenge of generating a strong
recovery and the longer-run challenge of rebalancing the economy.

[

The Determinants of Saving
To understand the behavior of consumption, it is critical to consider
how households divide their disposable income between consumption and
saving. Figure 4-2 shows the personal saving rate (that is, the ratio
of saving to disposable personal income) since 1960 (left axis), along
with the ratio of household wealth to disposable personal income
(right axis).

[


The big swings in wealth reflect asset market booms and busts. Much
of the drop in wealth in the early 1970s reflects the stock market
decline associated with the first oil price shock.
The stock market booms of the mid-1980s and the late 1990s are
obvious, as is the decline in stock prices in the early 2000s. The
wealth decline in 2008-09 was the largest such experience in the
sample, reflecting large contributions from falling house prices as
well as stock prices.
Paralleling the behavior of the consumption-output ratio, the saving
rate showed no strong trend before roughly 1980. But it has shown a
marked downward trend since then. Economic theory suggests a variety
of factors that should influence saving, most notably changes in the
demographic structure of the population, the growth rate of income,
and the real after-tax interest rate. None of these three factors,
however, provides a compelling explanation for the fluctuations in
the saving rate evident in the figure.
Indeed,some of the factors should probably have pushed saving up in
recent decades,not down. A 1991 study,for example,predicted that the
saving rate would rise as the baby boom generation entered its
high-saving pre retirement years  (Auerbach, Cai, and Kotlikoff 1991).
Instead, the saving rate fell steadily as the boomers approached
retirement (the first boomers claimed early Social Security benefits
in 2008).
Figure 4-2 suggests to the eye, and statistical analysis confirms,
a strong negative association between the saving rate and the wealth-
to-income ratio. This relationship has been interpreted as reflecting
the effect of wealth on spending: a run-up in wealth leads to less
need for saving. Such an interpretation is unsatisfying, however,
because it leaves a key question unanswered: If wealth movements
cause saving rate movements, what causes wealth movements? More
broadly, it leaves open the possibility that both saving choices and
asset price movements are a consequence of some deeper underlying
force. For example, an increase in optimism about future economic
conditions might lead both to a spending boom and to a general bidding
up of asset prices. In that case, the true moving force would not be
wealth changes per se; instead, both asset prices and saving would be
responding to the increase in optimism.
Survey data measuring ``consumer sentiment'' or ``consumer
confidence''do,infact,have substantial fore casting power for near-
term spending growth, and are also associated with contemporaneous
movements in asset prices (Carroll, Fuhrer, and Wilcox 1994). Such
surveys are therefore a useful part of a macroeconomist's fore
casting tool kit. But such surveys have not proven useful in
explaining long-term trends like the secular decline in the saving
rate.
Emerging economic research suggests another underlying explanation
that may be more potent: movements in the availability of credit. A
substantial academic literature has documented the expansion of
credit since the era of financial liberalization that began in the
early 1980s (Dynan2009). Many factors have contributed to this
expansion;perhapsthe most prominent explanation (aside from the
liberalization itself) is the telecommunications and computer
revolutions, which together have permitted the construction of
ever-more-detailed databases on consumer credit histories, giving
creditors a far more precise ability to tailor credit offers to the
personal characteristics of individual borrowers (Jappelli and
Pagano 1993). A beneficial effect of this information revolution has
been that many people who had previously been unable to obtain credit
have for the first time been able to borrow to buy a home,to start a
business,or to undertake many other useful activities (Edelberg 2006;
Getter 2006).
A reduction in saving, however, is almost the inevitable consequence
of a general increase in the ability to borrow. If there is less need
to save for a down payment for a home, for a child's education, for
unforeseen emergencies, or for spending of any other kind, then the
likelihood is that less saving will be done. Of course, eventually the
saving rate should mostly recover from any dip caused by a one-time
increase in the availability of credit, because whatever extra debt
was incurred must be paid back over time (and paying back debt is
another form of saving). This recovery in saving, however, may take a
long time. If, in the meantime, credit availability increases again,
the gradual small increase in saving that reflects debt repayment
could easily be obscured by the new drop in saving occasioned by the
continuing expansion in credit availability.
How much of the decline in the saving rate was due to a gradual, but
cumulatively large, increase in credit availability is not easy to
determine, partly because an aggregate measure of credit availability
is difficult to construct. Recent research on commercial lending has
argued that a good measure of the change in credit supply is provided
by the Federal Reserve's Senior Loan Officer Opinion Survey on Bank
Lending Practices, in which managers at leading financial institutions
are asked for their assessments of credit conditions for businesses
(Lown and Morgan 2006). Building on that research, one study has
proposed that a measure of the level of credit availability to
consumers can be constructed simply by accumulating the sequence of
readings from this survey's measure of credit availability to
consumers (Muellbauer 2007).ï¿½1A\1\
ï¿½1A\1\ï¿½1ASpecifically, each quarter the survey asks about banks'
willingness to make consumer installï¿½ment loans now as opposed to
three months ago.

Economic theory suggests that one further element may be important in
understanding spending and saving choices around times of recession:
the intensity of consumers' precautionary motive for saving. Because
the risk of becoming unemployed is perhaps the greatest threat to most
people's future financial stability, the unemployment rate has
sometimes been used as a proxy for the intensity of the precautionary
saving motive.

Implications for Recent and Future Saving Behavior
Figure 4-3 shows the relationship between the measured saving rate and
a simple statistical model that relates the saving rate to the wealth-
to-income ratio, a slightly modified version of Muellbauer's credit
availability index, and the unemployment rate. The statistical model
is estimated over the sample period 1966:Q3 to 2009:Q3. All three
variables have statistically important predictive power, with the two
most important measures being the measure of credit conditions and the
wealth-to-income ratio.

[


Figure 4-4 uses this simple framework to ask what the path of the
saving rate might have looked like if the increase in credit
availability and the housing price boom had not occurred.
(To be exact, the figure shows what the model says the saving rate
would have been if the wealth-to-income ratio had remained constant
from the first quarter of 2003 to the fourth quarter of 2007, and if
credit conditions had neither expanded nor contracted; the first
quarter of 2003 is chosen as the starting point because in that
quarter the wealth-to-income ratio was close to its average
historical value.) In this counter factual history,the personal
saving rate would have been,on average, about 2 percentage points
higher over the 2003-07 period.
Of course, a far more important consequence than the higher saving
rate might have been the avoidance of the financial and real
disturbances caused by the housing price boom and subsequent crash.
But taking the crash as given, Figure 4-3 shows that the model does a
reasonably good job in tracking the dynamics of the saving rate over
the period since the business cycle peak. All three elements of the
model contribute to the model's predicted rise in the personal saving
rate over the past couple of years: the increase in the unemployment
rate, the sharp drop in asset values evident in Figure 4-2, and the
steep drop in credit availability as measured by the Senior Loan
Officer Opinion Survey.

[

The saving model also has implications for the future path of
spending. Because of the important role it finds for credit
availability, the model suggests that the speed of the recovery in
spending is likely to be closely tied to the pace at which the
financial sector returns to health. This point underscores a chief
motivation for the Administration's efforts to repair the damage to
the financial system: a full economic recovery is unlikely until and
unless the financial system is repaired. The vital role that a healthy
financial sector plays in the functioning of the economy explains the
urgency with which the Administration has been pressing Congress to
pass a comprehensive and effective reform of the financial regulatory
system (see Chapter 6 for a detailed discussion of the
Administration's proposals).
Over a longer time frame, a resumption seems unlikely of the past
pattern in which credit growth persistently outpaces in come growth.
Instead, credit might reasonably be expected to expand,in the long
run, at a pace that roughly matches the rate of income growth.
Similarly, in keeping with the long-run stability of the wealth-to-
income ratio evident in Figure4-2, wealth plausibly might grow at
roughly the same pace as income-or perhaps a bit faster if investment
can sustain an increase in capital per worker. Finally, although
unemployment is likely to remain above its normal rate for some time,
it too can be expected to return to historically normal values in the
medium run. Under these conditions, the model suggests that the
personal saving rate will eventually stabilize somewhere in the range
of 4 to 7 percent, somewhat below its level in the 1960s and 1970s,
but well above its level over the past decade.
The saving rate has already risen sharply over the past two years
(which reflects an even steeper drop in consumption than in income).
As credit conditions and the unemployment rate return to normal, it is
plausible to expect a temporary partial reversal of the recent
increase, even if asset values do not return to their pre-crisis
levels. It would not be surprising, therefore, if the saving rate
dipped a bit over the next year or two before heading toward a higher
long-run equilibrium value. The prospect of temporary fallback in the
saving rate is also plausible as a consequence of the expected
withdrawal of some of the temporary income support policies that were
part of the stimulus package. On balance, however, the United States
seems now to be on a trajectory that will eventually result in a more
ï¿½normal,ï¿½ and more sustainable, pattern of household saving and
spending than the one that has prevailed in recent years.
While the underlying economic forces sketched here seem likely to
lead eventually to a higher saving rate even in the absence of policy
changes, the Administration has proposed a variety of saving-promoting
policy changes to enhance that trend over the longer term. These
include increasing the availability of 401(k)-type saving plans and
encouraging employers to gradually increase default contribution rates
(and to ensure that new employees' default saving choices reflect
sound financial planning). Economic research suggests that people
assume that if their employer offers a retirement saving plan, the
default saving ratein that plan probably reflects a reasonably good
choice for them, unless their circumstances are unusual (Benartzi
and Thaler 2004).


The Future of the Housing Market and Construction

The boom in construction spending that characterized the middle years
of the past decade made a substantial contribution to growth while it
lasted. When the residential investment engine began to sputter around
the middle of 2006, and then to stall, the ensuing correction in the
sector was correspondingly steep. With the benefit of hindsight, it is
now clear that much of the mid-decade's frenetic activity was based on
unsound financial decisions rather than sustainable economic
developments. As a consequence, construction has declined to below-
normal levels as the excesses work off. For the future, construction
activity is expected to pick up and contribute to the economic
recovery,although this activity is likely to be well below the very
high levels it reached in the mid-2000s.
The Housing Market
The residential investment boom can be measured in several ways. As
Figure 4-5 shows, new construction of single-family housing units
soared in the first half of the 2000s. Builders were constructing 30
percent more single-family housing units a year in the expansion of
the 2000s than in the 1990s boom. Housing investment as a share of
GDP averaged more than 5.5 percent over the 2002-06 period, compared
with an average of only 4.7 percent from 1950 to 2001. Figure 4-6
shows that from 1995 to 2005 the homeownership rate rose from 65
percent to 69 percent as mortgage underwriting standards loosened,
especially in the later part of the period.

[

It is now apparent that the mid-2000s level of new construction was
unsustainable. Analysis by the Congressional Budget Office (2008) and
Macroeconomic Advisers (2009) suggests the mid-2000s pace of starts
was well in excess of the underlying pace of expansion in demand for
new housing units based on house hold formation and other demographic
drivers.

[

The boom was followed by an equally dramatic bust. From their peak
in the third quarter of 2005 to the first quarter of 2009,single-
family housing starts fell by more than a factor of four. The
homeownership rate reversed course,and by the second quarter of 2009
had returned to its 2000 level. The share of housing investment in
GDP plummeted to 2.4percent in the second quarter of 2009.
Just as the mid-decade's high levels of construction and housing
market activity were not sustainable, the recent extremely low levels
of construction will not persist indefinitely. In 2009, housing starts
and the share of housing investment in GDP were well below their
previous historï¿½ical lows. In the long run, sounder underwriting
standards will require more would-be homeowners to take time to save
for a down payment before buying a home, suggesting that the
homeownership rate will ultimately settle at a level lower than its
recent peaks. Nonetheless, as the populaï¿½tion grows and the housing
stock depreciates, new residential construction will be required to
meet demand. The analyses by the Congressional Budget Office (2008)
and Macroeconomic Advisers (2009) suggest that the underlying
demographic trend of household formation is consistent with growth in
demand of between 1.1 million and 1.3 million new single-family
housing units per year, more than double the pace of single-family
housing starts in November 2009. Indeed, since the second quarter of
2009, housing construction has already rebounded a bit, making its
first positive contribution to GDP growth in the third quarter of 2009
since the end of 2005. But, as described in Chapter 2, the stocks of
new homes and existing homes for sale, vacant homes that are not
currently on the market, and homes that are in the process of
foreclosure and that are likely to be put on the market at some point
remain high. As a result, construction demand is likely to rise to its
long-run level only gradually while some demand is met by the stock of
existing units.
In short,as the housing market stabilizes and returns to a more
normal condition, its role as a major drag on economic growth seems
to be ending, and it is likely to contribute to the recovery. But
residential construction cannot be expected to be the engine for GDP
growth that it was during the housing boom of the mid-2000s.

Commercial Real Estate
The market for commercial real estate has also suffered in the
recession. Commercial real estate encompasses a wide range of
properties, from small businesses that occupy a single stand-alone
structure to large shopping malls owned by a consortium of investors.
Problems in the commercial real estate sector are less obviously a
result of overbuilding than those in the residential sector; instead,
they reflect the sharp decline in demand for commercial space and the
overall decline in the economy. The value of commercial real estate
increased notably between 2005 to 2007, spurred by easy credit
conditions, as measured for example in the Senior Loan Officer
Opinion Survey. By the end of 2004,the netnumber of banks reporting
they had eased lending standards for commercial real estate loans was
persistently larger than at any point in the history of the series.
Most banks did not begin tightening standards again until the end of
2006. The relative quantity of financing also increased over this
period; the ratio of the change in the value of commercial real
estate mortgages to new construction, which should increase when
debt financing becomes relatively attractive, reached a 45-year high
in 2003 and then continued to climb, peaking at the end of 2005 at
more than three times the historical average ï¿½1A/2/
ï¿½1A/2/ The numerator of the ratio is the seasonally adjusted change in
commercial and multifamily
residential mortgages (Federal Reserve, Flow of Funds Tables
F219 and F220). The denominator is seasonally adjusted construction
of commercial and health care structures, multifamily structures,
and miscellaneous other nonresidential structures (Department of
Commerce, Bureau of Economic Analysis, National Income and Product
Accounts Table 5.3.5). The median of the ratio from 1958 to 2000
is 0.46, while the 2005:Q4 value is 1.50.
In the nonresidential sector, high prices did not translate into a
dramatic increase in new construction (Figure 4-7). Rather, existing
owners of nonresidential properties used the cheap financing and price
increases to refinance or sell. Several factors appear to have played
a role in limiting new investment in this sector. First, a close
look at Figure 4-7 shows that nonresidential construction has
historically exhibited much less volatility than residential
construction, a pattern that also held true during the recent boom.
Second, developers seem to have been wary of overbuilding because of
unhappy experiences in previous expansions. A final dampening factor
has been that construction resources were tied up in the residential
construction sector. Indeed, only when residential construction
slowed in 2006 did nonresidential construction begin to show larger
gains.

[

Commercial real estate values have declined dramatically since 2007.
As Figure 4-8 shows, according to the Moody's/REAL Commercial Property
Index, which tracks same-property price changes for commercial office,
apartment, industrial, and retail buildings, commercial real estate
prices fell 43 percent from their peak in October 2007 to September
2009. A steep increase in vacancy rates, stemming from weakness in the
overall economy, has been one important reason for these declines in
value: the commercial real estate services firm CB Richard Ellis
reports that vacancy rates for offices increased from 12.6 percent in
mid-2007 to 17.2 percent in the third quarter of 2009. Before the
recession, vacancy rates were generally declining.

[

As commercial real estate values have declined, owners have found it
difficult to refinance their debt because loan balances now appear
large relative to the properties' value. Nearly half of the banks
responding to the Senior Loan Officer Opinion Survey in the third
quarter of 2009 reported that they continued to tighten standards on
commercial real estate loans, whereas none of the respondents reported
having eased standards. Since commercial real estate loans typically
are relatively short term, an inability to refinance debt has led to a
sharp rise in delinquencies and foreclosures. Figure 4-8 shows that
the proportion of commercial real estate loans with payments at least
30 days past due rose from about 1 percent during most of the decade
to almost 9 percent by the third quarter of 2009. Distress has made
lenders reluctant to provide financing for new projects. Overall, the
value of commercial and multifamily residential mortgages declined in
each of the first three quarters of 2009 (Federal Reserve Flow of
Funds Tables L.219 and L.220). Tight credit and the increase in sales
of distressed properties have fed into further price declines,
generating a negative feedback loop between property values and
conditions in the sector.
As private sources of funding have dried up, the Federal Reserve has
helped fill the gap through the Term Asset-Backed Securities Loan
Facility (TALF). In June 2009, the TALF made lending available to
private financial market participants against their holdings of
existing commercial mortgage-backed securities (CMBS), thereby
increasing liquidity in the CMBS market. In November 2009, the TALF
made its first loans against newly issued CMBS. The provision of TALF
financing for these newly issued securities may prove particularly
important in allowing borrowers to refinance.
The negative feedback loop between credit conditions, the sale of
distressed commercial properties, and commercial property values may
lead to further price declines. Eventually, however, a combination of
economic recovery and an improvement in financing conditions should
help prices stabilize. Still, as with the residential mortgage market,
commercial real estate financing will likely not return any time soon
to the easy terms that prevailed before the collapse. Experience in
previous business cycles suggests that recovery of the sector will lag
the economy as a whole.

Business Investment

If consumption and construction are not the drivers of growth going
forward in the way they were in the early 2000s, two components of
private demand are left to fill the gap: business investment excluding
structures, and net exports ï¿½1A/3/ Nonstructures investment could well
become again (as it was in the 1990s) a driving force in the expansion
of aggregate demand and economic production. And in the long run, its
share in GDP could reach levels higher than those of the first part of
the decade.
ï¿½1A/3/ In the National Income and Product Accounts, construction of
commercial structures is classified as part of business investment.
Given that the boom and bust were concentrated in residential and
commercial construction, however, for discussing recent and
prospective developments it is more useful to consider commercial
construction investment together with residential investment, as was
done in the previous section. Thus, the discussion that follows is
largely concerned with nonstructures investment.

Investment in the Recovery
Investment spending (other than structures) plummeted in late 2008 and
early 2009. This investment spending fell so low that, after
accounting for depreciation, estimates of the absolute stock of
capital showed stagnation in 2008 and even a decline in the first
quarter of 2009. Falling spending in this category reflected falling
business confidence, as indicated, for example, in the Federal
Reserve Bank of Philadelphia's Business Outlook Diffusion Index;
this index was negative every month from October 2008 to July 2009,
signaling that more businesses thought conditions were deteriorating
than thought they were improving. Similarly, the National Federation
of Independent Business Index of Small Business Optimism hit its
lowest point since 1980 in March 2009.

Investment of this kind firmed in the second half of 2009, coinciding
with improvements in business confidence. Indeed, investment in
equipment and software increased at a 13 percent annual rate in the
fourth quarter. Nevertheless, the cumulative erosion has been so
substantial that years of strong growth will be necessary to fully
recover from the nadir. As a result, recovery of spending in this area
is likely to make a substantial contribution to the recovery of the
overall economy.

Investment in the Long Run
In the long run, the share of business investment is likely not just
to return to its pre-recession levels, but to exceed them. During the
boom of the 1990s, the share of business investment in equipment and
software as a fraction of GDP rose from a post-Gulf-War recession low
of 6.9 percent in 1991 to 9.6 percent in 2000. During that period,
investment in information processing equipment and software made the
largest contribution to the increase, as shown in Figure 4-9.
Information technology (IT) investment grew an astounding 18 percent
per year on average from 1991 to 2000. Other investment in equipment
and software, which includes industrial, transportation, and
construction equipment,accelerated as well,and grew as a share of GDP
over this period. This high level of investment in the 1990s
increased industrial capacity by an average of 4 percent per year.
As the figure shows, the boom came to an end at the beginning of the
2000s, when investment in every category of equipment and software
fell sharply as a share of GDP. The recovery in business investment in
equipment and software after the 2001 recession was weak. IT
investment grew at a historically tepid pace of 6 percent per year
from 2003 to 2007, far below pre-2000 growth rates. Non-IT investment
growth was also muted, with spending on industrial equipment growing
at an annual pace of only 3.7 percent from 2003 to 2007, down from
an average of 5.4 percent in the 1990s. Investment in transportation
equipment surpassed its 1999 peak only for one quarter in 2006.
In the recovery following the 2001-02 recesï¿½sion, the peak value of
non-IT equipment investment as a share of GDP was only 4.3 percent
(in 2006), a level that does not even match the historical average
value of that series in the period from 1980 to 2000. Production
capacity in the sector grew an average of 0.6 percent per year from
2003 to 2007, substantially below the average pace of growth in the
1990s. Taken as a whole, these figures suggest that business
investment may have been abnormally low over the course of the
post-2001 expansion.
There are strong reasons to expect investment's role in the economy
will be larger in the future. In the long run, the real interest rate
will adjust to bring the demand for the economy's output in line with
the economy's

[
capacity. The increase in private saving described in the first part
of the chapter, together with the policies to tackle the long-run
budget deficit that are the subject of the next chapter, should help
maintain low real interest rates. By keeping the cost of investing
low, these low real interest rates should help to encourage
investment.
At the same time, other forces should help increase investment at a
given cost of borrowing. A number of promising technological
developments offer the prospect that businesses will be able to find
many productive purposes for new investments, ranging from new uses of
wireless electromagnetic spectrum, to new applications of medical and
biological discoveries opened up by DNA sequencing technologies, to
environmentally friendly technologies like new forms of production
and distribution of clean energy (see Chapter 10 for more
on these subjects).
Another form of investment is business spending on research and
development (R&D). Such spending can be interpreted as investment in
the accumulation of ``knowledge capital.'' Ideally, private investments
in R&D will dovetail with complementary public investments in
knowledge capital through basic research and scientific and
technological infrastructure. The Administration's commitment to
fostering the connections between public and private investments in
knowledge production has been strongly signaled in both the Recovery
Act and the President's fiscal year 2010 budget (Office of Management
and Budget 2009). The Recovery Act included $18.3 billion of direct
spending on research, one of the largest direct increases in such
spending in the Nation's history. In addition, more than $80 billion
of Recovery Act funds were targeted toward technology and science
infrastrucï¿½ture. The Administration's first budget proposed to double
the research spending by three key science agencies: the National
Science Foundation, the Department of Energy's Office of Science,
and the Department of Commerce's National Institute of Standards and
Technology. And to foster private sector innovation,the budget also
included thefull $74billion cost of making the research and
experimentation tax credit permanent in order to give businesses the
certainty they need to invest, innovate, and grow.
With reduced demand from consumption and housing tending to make the
real interest rate lower than it otherwise would be, and increased
investment demand from the many newly developing technologies and
incentives for R&D, a larger portion of the economy's output is likely
to be devoted to investment. And, because business investment
contributes not only to aggregate demand but also to aggregate supply
and productivity, a larger role for investment will create a stronger
economy going forward.


The Current Account

The picture of future growth in the United States described in the
previous sections depends less on borrowing and consumption than did
growth in the past decade. This view has important implications for
our interactions with other countries and the current account.
Determinants of the Current Account
The current account is the trade balance plus net income on overseas
assets and unilateral transfers like foreign aid and remittances. The
trade balance, or net exports, represents the bulk of the current
account and is responsible for a large majority of short-run movements
in it. To a first approximation, a current account deficit implies
that the trade balance is negative or, equivalently, that our exports
are less than our imports. At the same time, the current account
deficit must also be matched by the net borrowing of the United States
from the rest of the world. If we spend more than we earn, we must
borrow the money to do so. In the national income accounting sense,
the definition of the current account can be reduced to national
saving minus investment (plus some measurement error).
This accounting definition provides a description but not an
explanation of the drivers of the current account. One important
driver is the business cycle. As Box 4-1 explains, over the last 30
years, the U.S. current account deficit tended to be larger when the
economy was booming and unemployment was low. In a boom,investment
tends to rise and saving tends to fall, generating a current account
deficit. When the economy struggles, investment often falls and
saving often rises, generating a surplus (or a smaller deficit). In
countries that rely more on exports to drive their growth, an
acceleration in growth can be associated with a rising current
account surplus (or smaller deficit).
Current accounts do not need to be balanced in every country in every
year. At any point in time, countries may offer more investment
opportunities than their desired level of saving at a given interest
rate can fund, making them net borrowers, resulting in a current
account deficit. Other countries may have an excess of saving over
desired investment, making them net lenders (a current account
surplus). However, in the
----------------------------------------------------------------------
Box 4-1: Unemployment and the Current Account

The relationship between the level of unemployment and the current
account balance is complicated. People frequently argue that imports-
and specifically the current account deficit-displace U.S. workers and
generate higher unemployment. However, the main determinant of
unemployment in the short and medium runs is the state of the business
cycle. The scatter plot of the current account and the unemployment
rate since 1980,shown in the accompanying figure, displays a positive
relationship. Historically, a smaller current account deficit has
coincided with a higher unemployment rate. Both were being driven by
cyclical economic factors: in a recession, the current account balance
improved, and unemployment was high. In a boom, the current account
balance deteriorated, and unemployment was low. This usual pattern has
been at work in the current recession. The U.S. current account
deficit narrowed from 6.4 percent of GDP in the third quarter of
2006 to 2.8 percent of GDP in the second quarter of 2009. At the
same time, unemployment rose from 4.6 percent to 9.3 percent.

The relationship between unemployment and the current account balance
can be different in countries that have relied more heavily on exports
for growth. For example, in Germany, the unemployment rate fell from
11.7 percent in 2005 to 9.0 percent in 2007 while the current account
surplus rose from 5.1 percent of GDP to 7.9 percent. Likewise, in
Japan, unemployment fell from 2005 to 2007 as the current account
surplus rose. Given the slack in the U.S. economy, a shift toward a
current account surplus could increase aggregate demand and help lower
the unemployment rate.


[

---------------------------------------------------------------------
long run, current accounts should tend toward balance, thereby
allowing the net foreign investment position (total foreign assets
minus total foreign liabilities) of borrowing nations to at least
stabilize as a ratio to GDP and possibly to decline overtime.
Otherwise, credit or nations would be continually increasing the
share of their wealth held as assets of debtor nations, and debtor
nations would owe a larger and larger share of their production to
foreign lenders and capital owners.
Thus, in the long run, one would expect the U.S. current account to
move toward balance. As it does so, it will not cause the absolute
level of our accumulated net foreign debt to decline unless the U.S.
current account moves into surplus (which is of course possible). But,
even if the long-run current account is merely in balance or a small
deficit, the previous net foreign borrowing should still decline as a
share of GDP as GDP rises. Further, so-called ``valuation
effects''--changes in asset values of foreign assets held by Americans
or U.S. assets owned by foreign investorsï¿½also affect the ratio of
foreign indebtedness to GDP.
The Current Account in the Recovery and in the Long Run
As the U.S. economy recovers from the current crisis, it is unlikely
to return to current account deficits as largeas those in the
mid-2000s. Coming out of the 2001-02 recession, investment rose more
quickly than saving, and the current account deficit widened to more
than 6 percent of GDP (Figure 4-10). Investment had also declined
slightly more than saving had before the current crisis hit, and the
current account deficit moderated to less than 5 percent of GDP by
the third quarter of 2007 ï¿½1A/4/
ï¿½1A/4/There is also a statistical discrepancy between the saving-minus-
investment gap and the current account. While this discrepancy is
generally close to zero, it moved from slightly negative to slightly
positive in this period, so that the measured current account moved
more than the measured gap between saving and investment did.

The gap narrowed rapidly as investment fell sharply during the
crisis. The increase in the personal saving rate since the onset of
the crisis has partly offset the large Federal budget deficit
(which is negative government saving), so the current account
deficit shrank to under 3 percent of GDP.
The specific path of the current account as the economy exits the
crisis will depend on whether government and private saving rise
ahead of, or along with, a rebound in private investment. But in the
long run, the current account deficit is likely to be smaller than
it was before the crisis. The likely rise in private and public
saving relative to their pre-crisis levels

[

implies an increase in national saving. Thus, saving is likely to more
closely balance domestic investment, suggesting a transition to a
smaller current account deficit than in the 2000s. Given that the
current account deficit has already narrowed to roughly 3 percent of
GDP-less than half its peak-the crucial challenge will be to avoid a
reversion to a high-spending, low-saving economy. A successful shift
toward a more balanced world growth model generated by increased
consumption in nations with current account surpluses could improve
net exports even more. This could bring the current account deficit
toward its mid-1990s level of roughly 1 to 2 percent of U.S. GDP.
Exports can be expected to rise rapidly as the world economy
recovers for a number of reasons. Just as trade typically falls
faster than GDP in a recession (discussed in Chapter 3),it typically
grows faster during are bound. Trade-to-GDP ratios have fallen in
the last year and can be expected to bounce back as the world
economy recovers. This bounce-back alone will lead to rapid export
growth. More generally, the crucial driver of exports is always the
performance of the world economy. For U.S. goods and services to be
bought abroad, demand in other countries must return robustly. This
is one reason for the United States to strengthen its ties with
fast-growing regions such as emerging East Asia. The faster our trade
partners grow and the more we trade with fast-growing economies, the
more demand for U.S. exports grows. Figure 4-11 shows the historical relationship between U.S. export growth and growth of non-U.S.
world GDP.
The rebalancing of the U.S. economy is likely to be accompanied by a
rebalancing of the world economy as well. It is reasonable to expect
growth in East Asia to continue at a rapid rate but also to become
more oriented toward domestic consumption and investment than it has
been in the recent past. Some nations with large current account
surpluses took steps to increase domestic demand during the crisis,
and these efforts must be maintained and expanded if world growth is
to rebalance. It is not a given that such a transition in world demand
will take place. Concerted policy action will be needed, but if saving
falls in countries with current account surpluses and spending rises,
that should stimulate U.S. exports as well as take pressure off of the
U.S. consumer as an engine of world growth.

Steps to Encourage Exports
The Administration is taking many concrete steps to encourage exports.
The Trade Promotion Coordinating Committee brings governï¿½ment agencies
together to help firms export. While the final decision of whether and
how much to export is a market decision made by private businesses,
the government can play a constructive role in many ways. The

[

Export-Import Bank can help with financing; consular offices can
provide contacts, information, and advocacy; Commerce Department
officials can help firms negotiate hurdles; a combination of agencies
can help small and mid-sized businesses explore overseas markets. Much
of the academic literature in trade models a firm's decision to export
as involving a substanï¿½tial one-time fixed cost (Melitz 2003). The
Administration is doing all that it can to lower that initial fixed
cost to help expand exports.
In addition, the Administration is pursuing possible trade agreements
and making the most of its current trade agreements to expand
opportuniï¿½ties for American firms to export. Because U.S. trade
barriers are relatively low, new trade agreements often lower barriers
abroad more than in the United States, opening new paths for U.S.
exports. As the Administration works to expand U.S. market access
through a world trade agreement in the Doha round of multilateral
trade talks, it continues to explore its options in bilateral free
trade agreements and regional frameworks, such as the Trans-Pacific
Partnership. The United States Trade Representative continues to work
through previously negotiated trade agreements to lower non-tariff
trade barriers and facilitate customs issues to make it easier for
U.S. businesses to export.
Not all of these developments will necessarily increase net exports
(or the current account) of the United States. Since the current
account equals net lending to or borrowing from the world, moving the
current account balance requires adjustments in saving and investment
as well as more opportunities to export. In the long run, increases
in demand for U.S. exports resulting from export promotion or reduced
trade barriers will generate higher standards of living, but through
improved terms of trade, not an increase in net exports. Further, the
simple recovery of world trade volumes will increase exports and
imports alike. As discussed in Chapter 10, this increase in trade can
increase productivity and living standards, but it will not change
the current account. However, rapid world growth and declining
current account surpluses abroad should lead to an increase in U.S.
exports. This can help increase U.S. net exports and hence contribute
to the recovery.
As with higher investment, lower current account deficits have
important long-run benefits. Lower foreign indebtedness than the
country otherwise would have had means reduced interest payments to
foreigners. Equivalently, it means that foreigners have on net smaller
claims on the output produced in the United States. Thus, lower
current account deficits will raise standards of living in the long
run.

Conclusion

Economic policy should not aim to return the economy to the path of
unstable, unsustainable, unhealthy growth it was on before the
wrenching events of the past two years. We should-and can-achieve
something better. Growth that is not fueled by unsustainable
borrowing, and growth that is based on productive investments, is
more stable than the growth of recent decades. And growth that is
associated with higher saving will lead to greater accumulation of
wealth, and so greater growth in our standards of living.