Error Correction in the Political Economy
In broad terms… (Fed) policy has been
generally formulated in
a forward looking manner with price
stability and economic
stability serving as implicit
or explicit guides.
Athanasios Orphanides
Historical
Monetary Policy Analysis and the Taylor Rule
(2003)
In the previous article, we suggested that the U.S. stock
market error corrects rather than mean reverts. The main cointegrating regression
is almost entirely descriptive over
the long-run. Why is this so?
Granger (1981) mentions three possible short-term error
correcting processes:
1) Arbitrage, that is
the buy and selling of an identical asset in separate markets. This concept is
appropriate for short-term trading rather than investing; it is unlikely to be
a factor over the long-term.
2) The input and
output of a process with memory, such as the cars entering and leaving the
Lincoln Tunnel.
3) Optimal forecasts
when the differenced variables are Gaussian.
The long-term error correction that we have measured in
the U.S. stock market over a period of years is not likely to originate from
any of these three sources. Where does the cointegration between the
willingness of investors to buy stocks and future industrial capacity
utilization come from? This article discusses the responses of the wider
society to a changing world.
The political system of the United States is not often
thought of as error correcting. Yet the world’s first large-scale democracy has
survived and prospered, not only because of the values upon which the nation
was founded, but also because of the practical institutional arrangements that
enable the nation to respond appropriately to environmental challenges. The
U.S. political system works by balancing interest against interest or, in the
case of the markets, opinion against opinion.
The Declaration of Independence of 1776 states according
to Enlightenment values that the government was founded to secure “Life,
Liberty, and the pursuit of Happiness.” To secure these values, the Declaration
and the Constitution of 1787 incorporated three basic principles:
1) The Rule of Law
The Rule of Law is the major
source of a durable community consensus. The legal system is an arrangement of
shared power between referee judges, who administer the rules of contest, and juries
of citizens, who assess the objective facts from the arguments of the
interested litigants and apply these facts to relevant law. The Rule of Law
enables the settlement of differences under a regime of ordered liberty.
2) The Balance of
Powers
The founders of the Republic
sought to preserve the freedoms of the citizenry. A principal means was the
Balance of Powers, the division of government into three co-equal branches,
with the legislature itself further divided. However, the founders of the
government also recognized the need for effective action. In the Federalist
Papers, Madison (1788) wrote that the reason of the public emerges from the
checks and balances provided by the constitutional process.
To illustrate the rationality of
this process: You are deciding whether or not to buy a stock. List on a T
account all the pros and then all the cons and make your decision on the
balance.
A further balance of power
between the government and society results from a pluralist society comprised
of many different groups that can form a majority in the political system
described above. The error correcting argument for pluralism is that it allows
within society a wide range of reforming responses that are useful as the
external environment changes.
3) Popular
Sovereignty
The people are the ultimate
source of authority, which they exercise at election time.
How is this relevant to the
market system? Yearly error correction in the U.S. stock market accomplishes
this task only somewhat. Why does market error correction occur in the
long-term? We have described a set of interlocking societal checks and
balances. A major regulator of the U.S. stock market is the countercyclical
policy of the Federal Reserve. The Fed’s monetary policy modulates the economy,
thus increasing economic
predictability.
Major events, however, will vary
with each economic cycle. The regression therefore calculates an equilibrium
level that the market will eventually cross as the result of error correction,
rather than a more placid Gaussian mean that the market will quickly revert to,
absent events. The stock market is neither chaotic, nor is it precisely
Gaussian. The long-term willingness of investors to buy stocks is cointegrated
with a Fed modulated industrial capacity utilization.
The Federal Reserve system was
founded in 1913 to, among other things, conduct the nation’s monetary policy.
The resulting management process is definitely not Gaussian, also requiring
good judgment.* Consider Alan Greenspan’s July 20, 2004 semiannual “Monetary
Policy Report to the Congress.”
“Economic developments in the United States have generally been quite
favorable in 2004, lending increasing support to the view that the
expansion is self-sustaining…What does seem clear is that the concerns
about the remote possibility of deflation that had been
critical in the
deliberations of the Federal Open Market Committee (FOMC)
last year
can now be safely set aside…As we attempt to assess and manage these
risks, we need, as always, to be prepared for the unexpected and to
respond promptly and flexibly as situations warrant. But
although our
actions need to be flexible, our objectives are not. For
twenty-five
years, the Federal Reserve has worked to reestablish price
stability
on a sustained basis.”
Error correction requires both the Balance of Powers, providing advice
and
consent, and the prescience to chart the course.
|
* Can monetary policy be conducted according to the Taylor rule, which assumes that the relevant economic parameters are perfectly known at the time?
The Taylor rule states:
i = 2 + p + ½ (p- 2) + ½ (q-q*)
where: i is the nominal fed funds rate p is the rate of inflation (p-2) is the difference between actual and targeted inflation of 2% (q-q*) is the present output gap of the economy, stated in percents. If the equilibrium real interest rate is 2% and inflation is 2%, and there are no errors to correct, the nominal fed funds rate should be 4%.
Since the Fed considers important both economic growth and the control of inflation, deviations from economic output and the targeted inflation rate have equal weights. Orphanides (2003) writes that the Taylor rule, and its variants, are useful organizing devices to describe Fed policy, after the fact. However, the model tracked actual Fed policy most accurately during the late 1970s, encompassing the Great Inflation, but would not have prevented it. The largest deviations from this model appeared during the more favorable 1960s and during the Volker years.
|