before inflation moves above 5 percent at least. The 4.5 percent inflation
rate, on average, for the half century following the abandonment of the
gold standard is not necessarily the norm for the future. Nonetheless, it is
probably not a bad first approximation of what we will face.
An inflation rate of 4 to 5 percent is not to be taken lightly—no one will
be happy to see his or her saved dollars lose half their purchasing power in
fifteen years or so. And while it is true that such a rate has not proved economically
destabilizing in the past, an inflation projection in that range assumes
a generally benign impact of retirement of the baby boomers, at least
through the year 2030. As we have seen, today's relative fiscal quiescence
masks a pending tsunami. It will hit as a significant proportion of the nation's
highly productive population retires to become recipients of our federal
pay-as-you-go health and retirement system, rather than contributors to
it. Over time, unless this is addressed, it could add massively to the demand
for economic resources and heighten inflationary pressures.
Thus, without a change of policy, a higher rate of inflation can be anticipated
in the United States. I know that the Federal Reserve, left alone,
has the capacity and perseverance to effectively contain the inflation pressures
I foresee. Yet to keep the inflation rate down to a gold standard level
*Included in this period was the seemingly anomalous low-inflation, low-interest-rate period of
the early 1960s, which had many of the characteristics of today's global disinflation. As I noted
earlier, its cause was in a way similar to the aftermath of the cold war, in that it was noneconomic:
the delayed application for commercial use of gains in military technology during World
War II and the large backlog of invention built up during the 1930s.
482
TH E DE LPH IC FUTU RE
of under 1 percent, or even a less draconian 1 to 2 percent range, the Fed,
given my scenario, would have to constrain monetary expansion so drastically
that it could temporarily drive up interest rates into the double-digit
range not seen since the days of Paul Volcker. Whether the Fed will be allowed
to apply the hard-earned monetary policy lessons of the past four decades
is a critical unknown. But the dysfunctional state of American politics
does not give me great confidence in the short run. We could instead see a
return of populist, anti-Fed rhetoric, which has lain dormant since 1991.
My fear is that as Washington strives to make good on the implicit
promises made in the social contract that characterizes contemporary
America, CPI inflation rates by 2030 will be some AVz percent or higher.
The "higher" is meant to reflect whatever inflation premium might arise as
a consequence of the inadequate funding for health and retirement benefits
for baby boomers. In the end, I see a positive fiscal outcome, as I note in
chapter 22. But I suspect it is likely that to restore policy sanity we will first
have to trudge through economic and political minefields before we act
decisively. I am reminded of Winston Churchill's perception of Americans,
who "can always be counted on to do the right thing—after they have exhausted
all other possibilities." The trip through the minefields is a major
source of risk for my forecast, and it could be manifested in higher paths
for interest rates and inflation.
An elevated inflation rate, if allowed to develop, will create a different
financial environment than currently prevails. In part, this is because its
emergence is likely to parallel a decline in the developing world's currently
above-normal propensity to save. As noted earlier, developing-country
savings rates have historically averaged only a few percentage points higher
than those of developed countries. But owing to the combination of a surging
China, with its historically high savings rate,* and OPEC's recent huge
accumulation of liquid assets/ savings rates in developing countries ballooned
to 32 percent in 2006, while the rates in developed countries averaged
less than 20 percent.
*China's savings rate is the result both of low government provisions for health and retirement
and of soaring business savings.
tOil-exporting nations reported $349 billion in foreign exchange at the end of 2006, compared
with only $140 billion at the end of 2002.
483
THE AGE OF TURBULENCE
As China continues its trek toward Western consumerism, its savings
rate will fall. And though oil prices are more likely than not to go higher, any
increase in OPEC's savings rates is likely to be far less than has occurred
since 2001. Implicit in such a scenario is a consequent removal of an excess
of saving intentions over investment intentions and, therefore, the lifting of
that important factor that has helped suppress real interest rates since early
this decade. Moreover, having largely bestowed its benefits, globalization
will slow its pace. The recent frenetic pace of world economic growth will
decline. The World Bank estimates that annual global GDP growth at market
exchange rates will slow to 3 percent over the next quarter century.
Global GDP grew at a 3.7 percent annual rate between 2003 and 2006.
The dispersion of current account balances, a function of the pace of
the globalized division of labor and specialization, should also slow. The
U.S. current account is thus likely to shrink, though aggregate world imbalances
may not. Other countries could eventually replace the United States
as the major absorber of cross-border saving flows.
With real interest rates and expected inflation likely to rise on average
over the next quarter century, so would nominal long-term rates. The order
of magnitude of interest rate change is difficult to pin down because of the
uncertainties that a quarter century can bring. But for illustrative purposes,
if real rates on ten-year U.S. Treasury notes were to rise by 1 percentage
point from today's 2.5 percent (owing to a fall in global saving intentions)
and if fiat-money inflation expectations added the 4.5 percentage points it
has implied in the past, that would create a nominal yield for the ten-year
note of 8 percent. Again, this excludes whatever premium is required to
fund the obligations to baby-boomer retirees. But we can take this level as
illustrative: sometime before 2030 the world is likely to be trading ten-year
U.S. treasuries at a rate of at least 8 percent. This level is only a baseline—an
oil crisis, a major terrorist attack, or an impasse in the U.S. Congress over
future budget problems could send long-term rates significantly higher for
brief periods.
There are other threats to the long-term financial stability of the United
States and the rest of the world besides a rise in riskless interest rates. A hallmark
of the past two decades has been a persistent fall in risk premiums. It
is difficult to discern whether investors believe underlying risks have dimin
484
TH E DE LPH IC FUTU RE
ished and hence they do not require the yield premiums over riskless treasuries
that were prevalent in the past, or whether it is a need for additional
interest income that is pushing them to reach for higher-yielding debt instruments.
Spreads over U.S. treasuries of CCC-rated corporate bonds (socalled
junk bonds) in mid-2007 were mind-bogglingly low. For example,
this spread declined from 23 percentage points amid a plethora of junk bond
defaults at the end of the recession in October 2002 to little more than
4 percentage points in June 2007, despite a large rise in issuance of CCC
bonds. Spreads of emerging-market bond yields over those of U.S. treasuries
have declined from 10 percentage points in 2002 to less than IVz percentage
points in June 2007. This compression of risk premiums is global. I am
uncertain whether in periods of euphoria people reach for an amount of
risk that is at the outer limits of human tolerance, irrespective of the institutional
environment in which they live. The prevailing financial infrastructure
perhaps merely leverages this risk tolerance. For decades prior to the
Civil War, banks had to hold capital well in excess of 40 percent to secure
their notes and deposits. By 1900, national banks' capital cover was down to
20 percent of assets, to 12 percent by 1925, and below 10 percent in recent
years. But owing to financial flexibility and far greater sources of liquidity,
the fundamental risk borne by the individual banks, and presumably investors
generally, may not have changed much over that time period.
It may not matter. As I noted in my farewell remarks to the Federal Reserve
Bank of Kansas City's Jackson Hole Symposium in August 2005,
"History has not dealt kindly with the aftermath of protracted periods of
low risk premiums."
At a minimum, as riskless interest rates rise and risk premiums are
purged of the unsustainable optimism they now embody, prices of income-
earning assets will surely grow far more slowly than during the past six
years. As a consequence of the decline in long-term nominal and real interest
rates since 1981, asset prices worldwide have risen faster than nominal
world GDP in every year, with the exceptions of 1987 and 2001-2 (the
years of the dot-com bubble collapse). This surge in the value of stocks, real
estate deeds, and other claims on income-earning assets—that is, direct and
indirect claims on assets, whether physical or intellectual—is what I desig
485
THE AGE OF TURBULENCE
nate an increase in liquidity. These paper claims represent purchasing
power that can quite readily be used to buy a car, say, or a company.
The market value of stock and the liabilities of nonfinancial corporations
and governments is the source of investments and hence the creation
of liabilities by banks and other financial institutions. This process of financial
intermediation is a major cause of the overwhelming sense of liquidity
that has suffused financial markets for a quarter century. If interest rates
start to rise and asset prices broadly fall, "excess" liquidity will dry up, possibly
fairly quickly. Remember, the market value of an income-earning security
is its expected future income leavened by a discount factor that changes
according to euphoria and fear as well as more rational assessments of the
future. It is those judgments that determine the value of stock and other income-
earning assets. It is those judgments that determine how much wealth
a society has. Large manufacturing plants, office towers, even homes, have
value only to the extent that market participants value their future use. If
the world were to come to an end in an hour, all symbols of wealth would
be judged worthless. Something far short of doomsday—say, a dollop more
of uncertainty added to the mix of our future outcomes—and market participants
will lower their bids and will value real assets less. Nothing has to
be happening outside our heads. Value is what people perceive it to be.
Hence liquidity can come or go with the appearance of a new idea or fear.
A related concern in financial markets is the large and continuing accumulation
of U.S. Treasury securities by foreign central banks, mainly in Asia.
Market participants fear an impact on dollar interest and exchange rates if
and when those central banks stop purchasing U.S. securities or, worse, try
to sell off large blocks of holdings. The accumulations are largely the result
of endeavors mainly by China and Japan to suppress their exchange rates
to foster exports and economic growth. Between the end of 2001 and
March 2007, China and Japan combined accumulated $1.5 trillion of foreign
exchange, of which four-fifths appears to be in dollar claims—that is,
holdings of U.S. Treasury and agency securities and other short-term claims,
including Eurodollars.*
*China has embarked on an announced program to diversify part of its huge foreign-exchange
reserves (1.2 trillion in dollars and the dollar equivalent of nondollar assets).
486
TH E DE LPH IC FUTU RE
Should the rate of accumulation slow or turn to liquidation, there will
surely be some downward pressure on the U.S. dollar exchange rate and upward
pressure on U.S. long-term interest rates. But the foreign-exchange
markets for the major currencies have become so liquid that the currency
transactions required to implement large international transfers of U.S. dollar
deposits can be accomplished with only modest disturbance to markets. For
interest rates, the extent of a rise is likely to be less than many analysts fear,
certainly less than a percentage point and conceivably much less. Liquidation
of U.S. Treasury securities by central banks (or any other market participant)
does not change the total amount outstanding of U.S. Treasury debt. Nor
does the outstanding amount of securities or other assets that the central
banks purchase with the proceeds of their sales. Such transactions are swaps,
which affect the spread between two securities but need not affect the overall
level of interest rates. It is similar to an exchange of currencies.*
The impact on interest rate spreads of a swap involving a large block of
U.S. treasuries by a central bank (or anyone else) depends on the size of the
portfolios of the world's other investors, and, importantly, the proportions
of those investments that are close substitutes of treasuries with respect to
maturity, the currency of denomination, liquidity, and credit risk. Holders
of close substitutes such as AAA corporate bonds and mortgage-backed securities
can be induced to swap for treasuries without undue disturbance
to markets.
The international financial market has become so large and liquid* that
sales of tens of billions of U.S. treasuries, perhaps hundreds of billions, can
be transacted without crisis-causing shocks to markets. We have had much
evidence of the market's capability to absorb major transfers of U.S. treasuries
