Thida also contributed this article from the Economist 25-Sep-99

A new economy for the New World: Inflation may not be as dead as it seems

ONCE upon a time sailors went to
sea in the belief that the earth was flat, putting them at risk of dropping
off the edge. Fortunately the discovery that it was round solved that
problem. Economic thinking about monetary policy and inflation has undergone
a similar about-turn.
In the 1960s, the conventional economic wisdom was that monetary policy
could reduce unemployment. The theoretical underpinning for this was the
Phillips curve, named after Bill Phillips, an economist from New Zealand
based at the London School of Economics. Mr Phillips, also a trained
engineer, constructed a machine to demonstrate the workings of the economy,
using water to represent liquidity. In 1958 he published a study showing
that between 1861 and 1957 some kind of trade-off between wage inflation and
unemployment seemed to have been operating in Britain: when unemployment was
high, inflation was low, and vice versa. This seemed to suggest that central
banks could permanently reduce unemployment by tolerating a bit more
inflation.

A decade later, however, two American economists, Milton Friedman and Edmund
Phelps, challenged this theory. The trade-off between inflation and
unemployment, they argued, was only short-term, because once people came to
expect higher inflation, they would demand higher wages, and unemployment
would rise back to its "natural rate", which depended on the efficiency of
the labour market. There was no long-term trade-off between inflation and
unemployment: in the long run, monetary policy could influence only
inflation. If policymakers tried to hold unemployment below its natural rate
(also known by an acronym, NAIRU , the non-accelerating inflation rate of
unemployment), inflation would be pushed ever higher.

Just as Messrs Friedman and Phelps had predicted, the level of inflation
associated with a given level of unemployment rose through the 1970s, and
policymakers had to abandon the Phillips curve. Today there is a broad
consensus that monetary policy should focus on holding down inflation. But
this does not mean, as is often claimed, that central banks are "inflation
nutters", cruelly indifferent towards unemployment.

If there is no long-term trade-off, low inflation does not permanently choke
growth. Moreover, by keeping inflation low and stable, a central bank, in
effect, stabilises output and jobs. Don Brash, governor of the Reserve Bank
of New Zealand, explains how this happens, using 4. The straight line
represents the growth in output that the economy can sustain over the long
run; the wavy line represents actual output. When the economy is producing
below potential (ie, unemployment is above the NAIRU ), at point A ,
inflation will fall until the "output gap" is eliminated. When output is
above potential, at point B , inflation will rise for as long as demand is
above capacity. If inflation is falling (point A ), then a central bank will
cut interest rates, helping to boost growth in output and jobs; when
inflation is rising (point B ), it will raise interest rates, dampening down
growth. Thus if monetary policy focuses on keeping inflation low and stable,
it will automatically help to stabilise employment and growth.

The Fed clearly understands this relationship between inflation and the
output gap, but initial signs suggest that the ECB may underestimate the
extent to which it can safely hold interest rates low in the short run to
boost output and jobs. It has repeatedly said that lower interest rates
cannot reduce unemployment. That is undoubtedly true in the long run, but if
an economy is operating below potential and the jobless rate is above the
NAIRU , then interest rates can safely be cut, and hence output boosted,
without pushing up inflation. Statements by ECB officials often appear to
suggest that all of the 10% unemployment rate in the euro area is structural
and none of it cyclical. However, estimates from the IMF suggest that output
in the area is currently almost 2% below potential, ie, the economy is
operating at point A . With growth in the euro area starting to pick up, the
ECB is already muttering about a rise in interest rates. But it is far too
soon: the area still has a significant output gap that will continue to hold
down inflation.

Is a bit of inflation good for you


When inflation was in double digits, central banks had a simple rule: bring
it down. But now that the rate in almost all rich economies is 2% or less,
they are being forced to ask: how low? This is hotly debated, because some
economists believe that inflation has economic benefits as well as costs.

Consider, first, the costs. When inflation is high, people find it difficult
to distinguish between changes in average prices and changes in relative
prices. For example, a firm cannot tell if a rise in the price of copper
reflects general inflation or a scarcity of the metal. This distorts
important price signals, leading to a misallocation of resources. Inflation
also creates uncertainty about the future, which reduces investment. And
lastly, because of the way inflation interacts with tax systems, which are
never fully indexed for inflation, it reduces the real return on saving and
hence reduces future growth.

Inflation in double digits clearly does considerable harm, but what about
lower rates? And does 5% inflation do more damage than 2%, or zero? Yes,
answers Martin Feldstein, president of the National Bureau of Economic
Research*. He believes that even modest inflation can do significant damage
through its effect on saving. Tax is levied on nominal interest income, so
as inflation rises, the real after-tax return on savings comes down and
people save less, which depresses future growth rates. Mr Feldstein
estimates that in America a cut in inflation from 2% to zero would
permanently increase the level of GDP by 1%. To reduce inflation by two
percentage points would involve a one-off loss in output of 5% of GDP , but
the discounted present value of the future annual gains, of around 35% of
GDP , would far outweigh the loss. Zero inflation, he concludes, is a worthy
goal.

Yet there is little empirical evidence that lower inflation rates do
noticeably improve growth performance once inflation is below 5%, say. A
scatter plot of inflation against growth for a range of countries shows no
clear trend (see 5). A study by Robert Barro, an economist at Harvard
University, found that a reduction in inflation of one percentage point
increases the annual growth rate by a paltry 0.02 of a percentage point.

Some economists go much further, arguing that modest inflation, of perhaps
3-4%, is good for growth and employment. Nominal wages, they say, tend to be
rigid downwards. Workers may be prepared to put up with flat wages when the
inflation rate is 3%, which amounts to a decline in real income, but they
are reluctant to accept a pay cut in money terms. So if the inflation rate
is zero, real wages cannot be adjusted downwards in declining industries or
regions, which means that unemployment will rise. Inflation, the argument
runs, greases the wheels of the labour market, allowing real wages to adjust
more smoothly.

A widely cited study by the Brookings Institution+, which examined pay in
America since 1959, confirmed that very few people take nominal wage cuts in
any year. However, the problem of wage "stickiness" may be overstated. There
have been few periods in the past when inflation has been less than 3% for
an extended period, so it is not surprising that falling wages are rare. If
inflation were to remain low, resistance to wage cuts might fade. Indeed, in
Japan wages have been falling over the past two years. And as long as
productivity is rising, allowing unit labour costs to fall, there may be no
need for nominal pay cuts anyway. America's recent experience certainly
suggests that nominal wage rigidities are no cause for concern: unemployment
has continued to fall even though average inflation over the past three
years has been only 2%.

A second common worry about zero inflation is that interest rates cannot
fall below zero, so there is no way of allowing real interest rates to
become negative to help the economy out of a recession. But the need for
negative real interest rates may be exaggerated. Mervyn King, a deputy
governor of the Bank of England, argues that negative real interest rates
have been rare in America during the past half-century?. During most
recessions, low real interest rates have been enough to boost demand. The
only time when there might be a problem is if the economy suffers a shock
when the output gap is already large and inflation below target because
policymakers have failed to react to a previous slump in demand. A
pre-emptive policy which aims to prevent inflation going below target as
well as above, argues Mr King, minimises the need for negative real interest
rates.

The final and most serious concern is that if central banks aim for zero
inflation, prices are more likely to fall for brief periods, and the
experiences of the 1930s and Japan today show that deflation can be more
dangerous than modest inflation. But falling prices are not necessarily a
problem. Before the first world war, a decline in the average price level
was quite common during periods of rapid technological change, such as the
late 19th century; yet these were also periods of strong growth. This is
quite different from the harmful sort of deflation seen during the Great
Depression. Moreover, the odd year of falling prices does not matter so long
as it does not feed expectations that they will continue to fall, causing
households to delay spending.

On balance, the benefits of retaining some inflation are probably
overstated. Price stability reduces uncertainty and so offers the best
economic environment for firms and households. So why do central banks not
invariably aim for zero inflation? The answer is that official
consumer-price indices tend to overstate the true rate of inflation in all
countries, because they do not adjust fully for improvements over time in
the quality of goods and services. Estimates suggest that in most countries
the official consumer-price index overstates inflation by around 0.5-2.0% a
year. This is the best reason why central banks should aim for a slightly
positive inflation rate.

Flat-earth economics


With rates of consumer-price inflation of 2% or under, most rich countries
have therefore more or less achieved "price stability". However, that does
not mean that central bankers can go out on the town and celebrate. Despite
subdued inflation, the big three central banks all face their own particular
dilemmas. Japan's economy is still at risk from deflation. The ECB is having
to deal with the added uncertainty of a completely new currency, which has
made the bank's monetary policy extra cautious at a time when output in the
euro area is stuck below its potential. As for the American economy, the Fed
is baffled because inflation is currently lower than any economic model
would have predicted.

Under the old economic rules, if unemployment fell below the NAIRU (thought
to be around 5.5%), inflation would start to rise. This, in turn, implied
that the maximum rate of growth the American economy could safely sustain
(adding up growth in the labour force and in productivity) was about
2.25-2.5%. Today America's jobless rate stands at a 30-year low of 4.2% and
GDP has grown at an average rate of almost 4% a year over the past three
years. According to the textbooks, inflation should be rising. Instead,
America's underlying inflation rate has remained subdued. Hence all the talk
about the Goldilocks economy-neither too hot nor too cold.

One popular explanation of why strong growth has not pushed up inflation is
that the American economy is undergoing a paradigm shift. In this "new
economy", it is argued, information technology and increased global
competition have opened up new investment opportunities and boosted
productivity growth. Mr Greenspan himself frequently sings the praises of
the IT revolution, and talks of a "once-a-century phenomenon". Other
economists go further and argue that in this new era of low inflation the
Fed can throw away its rulebook: the notions of a NAIRU and a maximum
sustainable rate of growth have become redundant. Inflation is dead.

Then again, perhaps not. America's productivity growth in the non-farm
sector has indeed increased, to an annual average of 2% over the past three
years, twice the average over the previous 25 years. The big question is
whether this is just a cyclical blip or a permanent increase in trend
growth. If (a big if) this increase is maintained, it may raise the economy'
s non-inflationary speed limit to 3% a year, say. But that does not mean
there are no limits to growth: if it continues at its recent pace and the
labour market continues to tighten, inflation is bound to rise.

Some people may want to believe in economic miracles, but there is an
alternative explanation for America's low rate of inflation. The country has
benefited from four favourable factors. First, weak oil and commodity prices
until earlier this year, and a strong dollar and weak overseas demand,
especially in Asia, have held down import prices. Second, an investment boom
has created excess capacity which has made it hard for firms to pass on wage
increases in higher prices. Although unemployment is at a 30-year low,
America's capacity utilisation is well below its historical average. Third,
non-wage labour costs have been curbed by a booming stockmarket, which has
allowed firms to reduce their contributions to employee pension plans. And
fourth, the inflation rate has been nudged down by changes in the way of
measuring it.

By holding down prices, these factors have in turn helped to restrain wages.
But the old relationship between a tight labour market and wage growth
remains alive and well: real wage growth has accelerated over the past few
years, as you would expect when the unemployment rate is below the NAIRU .
This may have been temporarily offset by the favourable factors, but as
these effects wear off, inflation is likely to creep up again. Oil prices
are already rising; the dollar has slipped; and recovery in the rest of the
world will push up import prices. There are also signs that nominal wage
costs are now starting to pick up. Goldilocks is in danger of burning her
tongue, and the Fed may come to regret that it did not raise interest rates
sooner. Despite the increases this summer, real interest rates still remain
lower than a year ago.

Reports of the death of inflation are therefore much exaggerated. Far from
being dead, inflation may have taken on a new, more dangerous guise.

*"The Costs and Benefits of Going from Low Inflation to Price Stability", by
Martin Feldstein. NBER working paper No 5469, 1997.

+ "The Macroeconomics of Low Inflation", by George Akerlof, William Dickens
and George Perry. Brookings Papers on Economic Activity 1, 1996.

? "Challenges for Monetary Policy: New and Old", by Mervyn King. Federal
Reserve Bank of Kansas City symposium, August 1999.


A note from David to fulfill my mission on name-dropping: Martin Feldstein
is another old boss.  George Akerlof is and Bill Dickens was a Berkeley
professor.  Feldstein is G.W. Bush's economic adviser, and would probably
be a President Bush's nominee for Secty of Treasury or to replace
Greenspan.  Akerlof is married to Janet Yellen.

David I. Levine                 Associate professor
Haas School of Business    ph: 510/642-1697
University of California    fax: 510/643-1420
Berkeley CA  94720-1900                            email:
levine@haas.berkeley.edu
http://web.haas.berkeley.edu/www/levine/