than a coin flip of the future direction of the price of a stock. But sometimes
that walk is interrupted by a stampede. When gripped by fear, people rush to
disengage from commitments, and stocks will plunge. And when people are
driven by euphoria, they will drive up prices to nonsensical levels.
So the key question remains, as I summarized it in a 1996 reflection I
shall never live down, "How do we know when irrational exuberance has
unduly escalated asset values, which then become subject to unexpected
and prolonged contractions?" It is often suggested that the richest investors
are those best at gauging shifts in human psychology rather than at forecasting
earnings per share of ExxonMobil. A whole school of stock-market
psychologists has arisen around this thesis. They call themselves Contrarians.
They trade on the view that irrational exuberance eventually ends up
in falling stock prices, as shares get bid up for no plausible reason, and then,
when that becomes evident, fear grips the market and prices unravel. Contrarians
pride themselves on trading against crowd psychology. Since stock
prices are cyclical, some do succeed by trading contrary to the crowd. But
you rarely hear about those who try this approach and lose their shirts. I
also never hear much from coin tossers who lose.
Perhaps someday investors will be able to gauge when markets veer from
the rational and turn irrational. But I doubt it. Inbred human propensities to
swing from euphoria to fear and back again seem permanent: generations of
experience do not appear to have tempered those propensities. I would think
that we learn from experience, and, in one sense, we do. I, for example, when
asked what worrisome imbalances and problems lie over the forecast horizon,
invariably respond that financial crises that are foreseeable by market
participants rarely happen. If a stock-market bulge is perceived to be the
precursor of a crash, speculators and investors will try to sell out earlier. That
defuses the nascent bubble and a crash is avoided. The sudden eruptions of
*I find the oft-quoted explanation that it was program trading unconvincing. As prices tumbled,
sellers could have turned off the program-trading switch.
466
TH E DE LPH IC FUTU RE
fear or euphoria are phenomena that nobody anticipates. The horrendous
decline in stocks on "Black Monday" came out of the dark.
Successful investing is difficult. Some of history's most successful investors,
such as my friend Warren Buffett, were early to understand the
now well-documented anomaly that the rate of return on stocks, even adjusted
for risk, exceeds that on less-risky bonds and other debt instruments,
provided one is willing to buy and hold equities for the very long run. "My
favorite holding period is forever," said Buffett in an interview. The market
pays a premium to those willing to endure the angst of watching their net
worth fluctuate beyond what Wall Streeters call the "sleeping point."
The lessons of stock-market investing apply to the forecasting of whole
economies. Because markets tend to steady themselves, a market economy
turns out to be more stable and forecastable over the long run than in the
short run—assuming, of course, that the society and institutions upon which
it rests remain stable. Long-term economic forecasting is grounded in two
sets of historically stable data: (1) population, which is the most forecastable
statistic with which economists deal, and (2) productivity growth, the consequence
of the incremental buildup of knowledge and the source of sustainable
growth. Since knowledge is never lost, productivity will always rise.*
What, then, can we reasonably project for the U.S. economy for, say,
the year 2030? Little, unless we first specify certain assumptions. I need affirmative
answers to the following questions to get started. Will the rule of
law still be firm in 2030? Will we still adhere to the principle of globalized
free markets, with protectionism held in check? (By protectionism, I mean
not just barriers to international trade and finance but governmental restrictions
against competition in domestic markets as well.) Will we have
fixed our dysfunctional elementary and secondary school systems? Will the
consequences of global warming emerge slowly enough so as not to significantly
affect U.S. economic activity by 2030? And finally, will we have kept
terrorist attacks in the United States at bay? Unsaid are those possibilities,
such as a wider war or a pandemic, that could upset any forecast. This is a
rather long list of preconditions, but unless I can assume them, it is futile to
venture very far over the horizon.
*Output per hour, the conventional proxy for productivity, can and does decline at times.
467
THE AGE OF TURBULENCE
In my experience, the most important is the nature of our rule of law.
I do not believe most Americans are aware of how critical the Constitution
of the United States has been, and will continue to be, to the prosperity of
our nation. To have had, for more than two centuries, unrivaled protection
of individual rights, and especially property rights, for all participants in our
economy, both native-born and immigrant, is a profoundly important contributor
to our adventuresomeness and prosperity. To be largely free of fear
of a secret police arbitrarily hauling us off for interrogation for "crimes" we
never knew existed is something not to be taken for granted. Nor is freedom
from the threat of arbitrary confiscation of a business to which we
have devoted much of our life. The principle of individual freedom touches
a deep cultural chord in Americans: the belief embodied in our Constitution
of the basic equality of all citizens before the law. Reality has not always
matched this ideal, and discrimination against African Americans in
particular forces us periodically to revisit the early constitutional debates
about slavery and its violent resolution in the Civil War; we've come a long
way, but we have a distance yet to travel.
America's unrivaled protection of property rights has long attracted
foreign investment to our shores. Some investors come in order to participate
in a vibrant, open economy; others simply view the United States as a
safe haven for their savings that is not available in their home country. As I
shall explain, the ability of the American legal system to extend those cherished
property rights to an economy predominantly driven by intellectual
property will be a major challenge. And, of course, most detrimental of all
to our standard of living would be a reemergence of protectionism and
other policies that seek stability by preventing the change that is necessary
for growth. Economic reregulation would be a distinct step backward in
our quest for a prosperous future.
The impact that fixing our school system would have on our future
levels of economic activity may not be easy to measure, but unless we do
so and begin to reverse a quarter century of increases in income inequality,
the cultural ties that bind our society could become undone. Disaffection,
breakdowns of authority, even large-scale violence could ensue, jeopardizing
the civility on which growing economies depend.
The timing of global warming's impact is even harder to foretell. To
468
TH E DE LPH IC FUTU RE
day's scientific consensus focuses on effects that are likely to emerge in the
second half of this century—a millisecond in climatological time but beyond
our forecast period. There is as yet little we can anticipate for the
years immediately ahead. Nonetheless, I would expect the markets to respond
even before the answers become clear—already insurers are rethinking
storm and flood coverage, for example. The prospect of climate change
is affecting energy markets as well.
Finally, there is the risk of a renewal of terrorist attacks. When engulfed
in fear, people disengage from the normal daily market interaction that is
an integral part of an economy based on the division of labor and specialization.
The terrorism of 9/11 was a defining moment that underscored the
critical value of our highly flexible, largely unregulated economy, which
weathered the shock with minimal longer-run consequences. We could
probably absorb terrorist attacks like those being experienced today in the
Middle East and Europe. But larger-scale attacks* or more widespread warfare
would surely be destabilizing.
I have been encouraged by the ability of market economies to persevere
through violence and the threats of violence. World Bank data indicate
that Israel has managed to create a per capita national income at nearly half
the level of that of the United States and roughly equal to per capita incomes
recorded in Greece and Portugal.+ Lebanon's GDP for 2006, despite
the confrontation between Hezbollah and the Israeli military, was down
only 4 percent. Even Iraq has managed to maintain a semblance of a functioning
economy through all its turmoil of recent years.
The long list of caveats does not inordinately tie our forecasting hands.
After all, such a list has always existed in one form or another, and yet the
record of long-term forecasting of the U.S. economy overall in my experience
has been reasonably impressive.
So given the presumed base of global, flexible markets protected by the
rule of law, what can we project as our most likely future? What is the most
likely level of overall activity we can expect in our arbitrarily chosen 2030
forecast year? We can project real GDP as long as we have projections of
*A nuclear detonation on U.S. soil, I fear, could temporarily unhinge our economy.
tU.S. aid accounts for only a small part of Israel's economy.
469
THE AGE OF TURBULENCE
hours worked and productivity, proxied by real GDP per hour. We know
with some degree of certainty the size of our population age sixteen and over
in 2030. Most of them have already been born. The proportion of the population,
especially the part under sixty-five years of age, that participates in
our labor force is high and reasonably stable. Our population older than
sixty-five years is expected to almost double by 2030, and the current 15
percent participation rate for that part of the labor force will rise as well, thus
adding a more than usual number of elderly workers by 2030. The size of
immigration within the politically and culturally feasible range of possibilities
does not matter much to the overall labor force forecast. Our next step
is to set the proportion of the total labor force that will likely be employed
in 2030 (or in a contiguous year if 2030 happens to be a year of recession).
Given our assumptions and the economy's historical record, it is difficult to
imagine the employment rate of the civilian labor force being outside the
rather narrow range of 90 to 96 percent (that is, an unemployment rate between
4 and 10 percent). America's fifty-year average is more than 94 percent,
with nonrecession years (the assumption for 2030) near 95 percent.
Combining labor force participation rates, population projections, a near
5 percent unemployment rate, and a stable workweek yields an annual growth
rate in hours worked in the United States through 2030 of 0.5 percent.*
The most encouraging aspect of productivity growth is how remarkably
stable it has been for the last century and more. Over much of that period, a
substantial boost in U.S. productivity reflected the shift of workers from
farms to urban factories and service establishments.1 But gains in national
productivity owing to farmworkers moving into higher-productivity nonfarm
jobs are essentially over. Less than 2 percent of the U.S. workforce
remains on farms, and that number is not likely to change much. Thus, future
national productivity growth will closely mirror the growth in nonfarm productivity.
Output per hour is the best measure we have of that growth.
*After a long decline from sixty-hour workweeks a couple of centuries ago, factory average
weekly hours settled at forty just after World War II and have held steady ever since. The shift
of the employment share to the service sector (where the workweek is shorter) has been reflected
in a slight overall decline in the weekly average.
tEven to this day output per hour on farms is less than in nonfarm regions, despite the remarkable
gains in crop and livestock yields achieved since the end of World War II.
470
THE DELPHIC FUTU RE
All gains in efficiency are the result of new ideas in the way people organize
their physical reality. To be sure, twenty-first-century human beings
are physically taller and stronger than earlier generations, thanks to improved
nutrition and health. But that has added very little to our ability to
produce. Over the generations it has been new ideas embodied in newly
built plant and equipment that better leverage and multiply human effort.
From the development of the textile loom two centuries ago to today's Internet,
output per hour has increased fiftyfold.
Statisticians usually attribute the growth in output per hour to three
primary economic "causes": the quantity of physical plant and equipment,
which they call "capital deepening"; the quality of labor input, which is a
reflection of education; and the otherwise unexplained, which they infer
results from organizational restructuring and new insights in how to generate
the nation's output. In all categories, productivity growth results from
ideas translated into valued goods and services. The quality of raw materials
used in the production process adds only modestly.
If we smooth through the raw data on output per hour, a remarkably
stable pattern of growth emerges, going back to 1870. Annual growth of
nonfarm business output per hour has averaged close to 2.2 percent since
then. Even without adjusting for the business cycle, wars, and other crises,
