|
INTERVIEWS
WITH ECONOMISTS
DR
WILLIAM COLEMAN
William
Coleman was a Research Officer at the Reserve Bank of Australia
in the early 1980s. He then completed his doctorate, on inflation
in the UK since 1870, at the London School of Economics. He has
since published on topics ranging across Central Bank independence,
exchange rate regimes, and the dependence of nominal interest
rates on inflation. He is currently a Senior Lecturer in financial
economics at the University of Tasmania.
Do you
agree that monetary policy should target one variable only? If
so, should this variable be inflation?
Monetary policy
should target just one variable. If you give it more than one
target you will just end up missing both targets.
Let me explain.
Think of a firm that has two targets: a target for revenue and
a target for sales volume. The firm can always set its price to
secure its revenue target, but if it does so it will miss its
volume target. Alternatively, it can set its price to reach its
volume target, but if it does so it will miss its revenue target.
‘Price policy’ alone cannot achieve both a revenue target
and a volume target. To achieve both targets the firm must bring
into play a second policy (e.g. advertising).
The same thing
applies to monetary policy; monetary policy can target only one
thing.
What should that target be? Inflation. Why? Because it is obvious
that monetary policy can affect inflation. The alternative target
is unemployment. Monetary policy does impact on unemployment in
the short run. But we are so ignorant of the size, timing and
duration of that impact that we can never know what is the appropriate
stance of monetary policy for any unemployment target. On top
of that, it is unlikely that monetary policy can affect unemployment
in the long run.
In brief,
monetary policy should stick to what it can do: stop inflation.
In your
view, what are the costs of unemployment and inflation? Which
of these costs is greater?
Unemployment
has the greater costs. If we could reduce unemployment by 5 per
cent that would add about $25 billion dollars to national income.
That is a significant sum: it is about twice the amount paid out
in old age pensions per year. It is equivalent to a bit under
$6000 per year for a family of four.
By contrast,
the direct costs of inflation are uncertain and, as far as they
can be made out, apparently very small. One direct cost you can
calculate is the ‘shoe leather’ cost of inflation; this
is the cost required of the more frequent liquidations of financial
assets, which take place under inflation as money holders seek
to minimise the amount of wealth they hold in the form of money.
One Australian estimate put this shoe leather cost of one per
cent inflation at two hundredth of one per cent of GDP! It is
trivial.
Do you
think the costs of making monetary policy too rigid are larger
than the costs of making it too variable?
The right
answer depends on whether we are concerned with the ordinary situation
or the exceptional event. My guess is that in the ordinary situation
the more costly error is being too variable. But it may be that,
in certain unusual events, being too rigid is the more costly
error. This may have been the case with the Great Depression,
for example, where a variable policy would almost have to have
done better than a rigid policy.
Perhaps I
can illustrate my position through a driving metaphor. I think
the road travelled by the economy is fairly straight and, as a
consequence, the ‘oversteering’ of variable policy is
more likely to cause problems than the ‘understeering’
of rigid policy. But, occasionally, there are sharp bends in the
road.
What is
your reaction to those advocating a policy target of zero inflation?
Would you give the RBA such advice?
Zero inflation
is overkill. One reason is that quality of products improves by
about 1 per cent per year. So a 1999 model car is about 1 per
cent higher in quality than a 1998 car. This means that with truly
no inflation its price would still rise by 1 per cent, simply
because you are getting ‘1 per cent more car’. So an
inflation rate of 1 per cent (as measured by the CPI) is in fact
like a true rate of no inflation.
If you
were governor of the RBA, what would you do that is different
from what is being done now?
The big challenge
to the RBA is to bolster the credibility of its goal of 2–3
per cent inflation. The trouble is, words are cheap; it is easy
to announce a goal of 2–3 per cent when inflation is within
2–3 per cent anyway. It is harder is to convince the public
that when inflation rises above 3 per cent the RBA will tighten
monetary policy, rather than change the goal. How does the RBA
convince the public of this? By proving to the public that the
RBA is willing to take the pain of tightening monetary policy.
The one effective way of proving that is for the RBA to have a
tough monetary policy — even when everyone else (journalists,
the public, politicians) think that inflation is beaten, and the
RBA should be loosening policy to reduce unemployment. The RBA
will make itself thoroughly unpopular by being so tough, and that
is the whole point of the exercise. The whole point is to prove
that the RBA is so set against inflation it is willing to incur
unpopularity in defeating it.
An analogy.
Why do gangs have initiation ceremonies? Because words are cheap;
because it is easy to say ‘I swear eternal allegiance to
a gang’. But to suffer the pain of an initiation ceremony
in order to join — that proves you mean it. The task of the
RBA today is to prove it ‘means it’ on inflation.
Do you
think that the central bank is capable of controlling the interest
rate or the money supply? If so, what is the extent of this capability?
The RBA is
perfectly capable of controlling the interest rate, but there
are some qualifications that must be kept in mind. First, it has
far more control over interest rates for short loans (say 90 days)
than long loans (say 10 years). Second, it has far more control
over nominal interest rates than real interest rates. Thirdly,
if the RBA does move the interest rate far from that rate paid
in other developed economies then the Australian dollar will depreciate.
The RBA tried
to control the money supply between 1975/76 and 1984/85, but it
abandoned the attempt in January 1985 when the money supply was
growing far faster than the target. It is not clear whether this
experience means that the RBA cannot control the money supply,
or whether the RBA did not try very hard to control it.
|