RESTORING FULL EMPLOYMENT:THE NATURAL RATE OF INFLATION VERSUS THE NATURAL RATE OF UNEMPLOYMENT

by     HAROLD CHORNEY,GRADUATE  PROGRAM IN PUBLIC POLICY AND PUBLICADMINISTRATION,DEPT. OF POLITICAL SCIENCE , CONCORDIA UNIVERSITY, MONTREAL

Paper presented to the Conference on Social Policy as if People Matter, Adelphi

University , Garden City , New York, Nov.12, 2004. Many thanks to Helen Ginsburg and Trudy Goldberg for their kindness in including me in this very interesting conference.This paper is also a preliminary version of a chapter in my forthcoming book.It contains an argument that I think is quite important as we struggle to escape from the damage of this long recession and restore prosperity. Adelphi also has an earlier version posted on their conference site.

RESTORING FULL EMPLOYMENT:THE NATURAL RATE OF INFLATION

VERSUS THE NATURAL RATE OF UNEMPLOYMENT

by     HAROLD CHORNEY,

GRADUATE  PROGRAM IN PUBLIC POLICY AND PUBLIC

ADMINISTRATION,

DEPT. OF POLITICAL SCIENCE , CONCORDIA UNIVERSITY, MONTREAL

Paper presented to the Conference on Social Policy as if People Matter, Adelphi

University , Garden City , New York, Nov.12, 2004.

INTRODUCTION:THE DEATH OF INFLATION AND THE RISE OF

UNEMPLOYMENT

For the past thirty years ever since the embrace by the central banks and the financial

markets of Milton Friedman’s counter-revolution in macro-economic theory

unemployment has averaged above 7 % in Canada and above 5 % in the United States.

Because of this many neo-classical economists who accept the doctrine of the natural

rate of unemployment argue that the natural rate lies between 7 and 8 % for Canada and

5 to 6 % for the United States.The natural rate argument turns on the notion that any rate

of unemployment below this rate is unstable and will cause inflation to accelerate

upwards eventually forcing unemployment up as well. I will deal at length with this

argument below.

For the time being simply note that  most other people who are not pure neo-

classical economists(for example: Keynesians, post Keynesians, institutionalists and

other heterodox economists as well as the general population)consider  these rates of

unemployment  far too high, imposing far too much hardship, poverty and loss of

potential output and income to be acceptable as a backdrop to a social policy as if people

mattered.

The Nobel prize winning economist, James Tobin argued in 1988 that it was

difficult to believe that the natural rate of unemployment could have risen to the level that

monetarists argued it had, particularly in Europe.

Why should economies that grew rapidly for twenty or thirty years with 2 or  3

% unemployment and low inflation suddenly have natural unemployment rates of  8,

9, or 10 per cent ? What institutional or structural changes …could have brought this

about ? (James Tobin, Monetary Policies in the 1980s and beyond.,in Full

Employment and Growth: Further Keynesian   Essays on Policy,

Cheltenham: Edward Elgar,1997)

I think Tobin was correct. There are no adequate explanations of structural or

institutional change that can possibly justify natural rates at these levels. In fact, in recent

years all of the changes have moved in the opposite direction: a smaller proportion of the

workforce that is unionized, less generous income replacement programs, fiercer

competition from low wage economies and easier mobility of capital due to increased

globalization.

My favourite joke about the natural rate of unemployment is that it is the rate of

unemployment above which the neo-classical monetarist and rational expectations

economists would lose their own jobs.

The point is that the natural rate  is a highly arbitrary rate about which there can

be considerable disagreement .Even the inventor of the concept, Milton Friedman

admitted in an interview with the Wall Street Journal in 1995  that “I don’t know what the natural rate is…and neither does anyone else.” (quoted in Amanda Bennett,”Business and

Academia Clash over a Concept:’Natural ‘Jobless Rate,” Wall Street Journal, Jan. 24,

1995,  A8. cited in R.Ehrenberg et al, Modern labour economics:theory and public policy,

Toronto: Pearson Education, 2004, p.516.)

During the period 1990 to 2001 in the US the unemployment rate averaged 5.5 %.

Since 2001 it rose above 6.0 % and stayed elevated above this level for a period of

months.Its current  October 2004 rate of 5.5% according to Joseph Stiglitz, the Nobel

prize winning economist underestimates the actual rate by as much as 3.5 percentage

points because of the large number of discouraged workers.(Full employment and Social

Well Being in a Global Economy, Keynote address, Adelphi University, Social Policy as if

People Matter, Conference, Nov.11, 2004. It is quite revealing  that when Alan Greenspan

relaxed his tight monetary policy during the second term of the Clinton years between

1997 and 2001 unemployment fell to as low as 4.0 % without any triggering of a

significant rise in inflation. The inflation rate peaked at 3.4% during this period.

Unfortunately in Canada monetary policy has been more dogmatic and the central

bank governor unwilling to relax his grip on rates more aggressively than the US, even

when the Canadian federal government was  running large surpluses which began in

1997.These surpluses have amounted to over $70 billion or close to 7 % of the Canadian

GDP.  For those years 1997-2001, the unemployment rate in the US averaged  4.5 %,

while Canada was mired in unemployment that averaged 7.8 %.

Unemployment rates in the two countries have tended to  follow each other

closely over the past 60   years with rates tending to be higher in Canada over the past

three decades. It is possible to break down the data into periods to get a better sense of

trends. One such data set has been assembled and calculated for the years 1947 to 1984

by Peter Sinclair and is incorporated  into Table one which partly extends the data set

until 2004. The trend to increasing unemployment after the switch in policy regime from

Keynesian to monetarist in both countries in the mid 1970s is clear.

It is also clear that prior to any adjustment for labour participation rates that

Canada has performed worse on the unemployment front than the United States from

1969 on.(Only since 2008 has the situation been reversed)  Prior to that from 1930 to

1958 its unemployment performance was superior in

all but two years. (See Appendix: Table one )

In fact from 1941 after having being sharply in excess of  10 %,  unemployment in

Canada fell to 4.4 % then to 1.4 % by 1944. For most of  the period 1931 to 1940

unemployment was severe.  The mean for this period was 13.1 %, the maximum 19.3 in

1933 and the minimum 9.1 in 1937. It remained below 5 % until 1958 when the first

serious post-war recession pushed it above 5 %. It returned to below 5 % by 1964 and

remained there until 1969 when it began its long upward rise.

The American experience during the depression years was also disasterous .(See

Appendix Table one) Unemployment rose to a peak of  22.3 % in 1932 and remained

above 15 % for five of  the years from 1931 to 1939.  The situation in Great Britain was

somewhat better. After having experienced very high unemployment during the late 1920s

unemployment reached 15.3 % by 1932 and averaged 10.9 %(maximum of 15.3 in 1932

and minimum of 5.8 in 1939)  throughout the decade of the thirties.

TABLE  ONE:

UNEMPLOYMENT CYCLES IN CANADA AND THE UNITED STATES 1945 -1984

US UNEMPLOYMENT                                                               CANADA UNEMPLOYMENT

Period    Mean    Maximum    Minimum  Relative to             Period Mean Max Min Rel.to                                                                     OECD average                                               OECD av.

__________________________________________________________________

1948-52    4.3         5.9             3.0              +1.2              1947-50  2.8    3.2   2.0 -0.8

1953-55     4.3         5.5             2.9             +08               1951-55   3.5    4.6  2.4  +0.3

1956-59     5. 2        6.8             4.1             +4.7              1956-58    5.0   7.0  3.4   +1.3

1960-61    6.1         6.7              5.5             +3.5              1959-65    5.6   7.0    3.9 +2.8

1962-68     4.1        5.7              3.6             +1.4              1966-68     4.1   4.8   3.5  +1.6

1969-72     4.9         5.8              3.4            +1.7              1969-73     5.6   6.2    4.4  +2.4

1973-78     6.5         8.3              4.8            +1.9              1974-78      7.1   8.3    5.3 +2.3

1979-84     7.8         9.5              5.8            +0.8              1979-84      9.4   11.3  7.4 +2.4

1984- 87    7.0         7.5             6.2                                  1984-86    10.5   11.3  9.6

1988-90     5.5        5.6              5.3                                   1987-90     8.0     8.1   7.8

1991-1992  7.2        7.5             6.8                                   1991-94     10.8   11.4  10.3

1993-1998   5.6       6.9             4.5                                   1995-98      9.1     9.6    8.3

1999 -2002  4.3       4.8             4.0                                   1999-02      7.4     7.6    6.8

2003-2004   5.8      6.1             5.4                                    2003-04      7.4    7.6    7.1

________________________________________________________________________

________

Sources: US Department of Commerce; UN Monthly Bulletin of Statistics; ILO; OECD

as cited in Table 1.10 in Peter Sinclair, Unemployment: Economic Theory and

Evidence,Oxford&New York:Basil Blackwell, 1987); US Bureau of Labour Statistics and

Statistics Canada.; Ontario Economic Outlook and Fiscal Review 2004.

During the same period in Europe, for the Euro group which followed the strict

monetarist guidelines of the European central bank and its stability pact the rate has

averaged above 8 % . Britain, Switzerland, Norway, Sweden  and Denmark  remain

currently  outside of the system. As might be expected their unemployment rate is

markedly lower. The rate for Britain, Sweden, Norway, Austria and Switzerland

currently, for example, averages 5.0 %.(See  table two below)

As a consequence of the last American  recession that began in the year prior to

9/11, unemployment is still 5.5 % in the United States and remains 7.1 % in Canada. (See

table two and table one in appendix  for the data) In countries like France, Germany and

Spain unemployment is 9.9  % and above.In Belgium it is over 13 %.  Only Britain

Austria,Switzerland and Japan currently have rates below 5 %.Over the period 1992 to

2004 unemployment in France has averaged 11.7 % . For the same period for Germany

unemployment has averaged  7.9 %, Italy 10.6%. For Japan, however the rate has

averaged 3.5% and for the UK 7.3 %.

While the current British rate is superior performance to Europe and Canada and a

major improvement over its performance in the first half of the 1990s it is still a marked

decline from what it managed in the years after the war. It is important to recall that

unemployment in Britain, for example averaged between 1 and 3 % during the period

1946 to 1960.(J.C.R.Dow,The Management of the British Economy, 1945-1960, fig.13.3,

p.342,Cambridge :Cambridge University Press, 1964)

Its unemployment performance deteriorated sharply under the monetarist policy

regime of Margaret Thatcher promoted by Sir Keith Joseph,  ably assisted by the

monetarist economists from the LSE like Alan Walters and  Harry Johnson and later  by

other leading monetarists like David Laidler and Michael Parkin from Manchester and

Terence Burns and  Alan Budd from the London Business School.( David Smith, The rise

and fall of monetarism, Harmonsworth:Penguin, 1987 pp.47-48.) Mrs. Thatcher came to

power in 1979 .

The impact upon unemployment is striking. Unemployment in Britain rose from

5.6 % in 1979 to 6.9 in 1980 to 10.6 in 1981. It continued to rise in 1982 to 12.3 %, to

13.1% in 1983 to 13.2% in 1984. It is true that inflation fell during this period from 13.4

% in 1979 to 5.0 % in 1984. But the cost in terms of lost output and hardship was close

to catastrophic.( Peter Hall, Governing the Economy:The Politics of State Intervention in  Britain and France, Cambridge: Polity press, 1986, p.120 Table 5.6. Unemployment data standardized OECD definition )

Peter Ham, an economist and special assistant to then Chancellor of the

Exchequer Dennis Healey put it insightfully in his study of the  Treasury in 1981.

The  recrudescence of vulgar monetarism in the late seventies has surely laid to

rest” the myth of    British pragmatism in policy matters.” Since 1974 a remarkable

degeneration in the formation of    British economic policy has taken place. This has

been closely associated with what has been    hailed by some as the ‘monetary

revolution’ and by others as a return to old values and ‘sound’    money.

The approach to economic management in the late 1970s and early 1980s more

closely resembled the application of the old fashioned Treasury rules of the

1920s than at any time in the fifty preceding years. The phenomena associated with

this violent throw back has been stagnation and decline…” (Adrian Ham, Treasury

Rules:Recurrent Themes in British Economic Policy, Quartet Books, London:, 1981,

pp.ix&1)

The apparent fact that both Kenneth Clark the Chancellor under Prime Minister

John Major(1992-1997) and to some extent Gordon Brown ,(1997 to present) the current

Chancellor under Tony Blair  moved some of the way back toward pragmatic

Keynesianism with good results ought to be more carefully investigated.

Of course, when making international comparisons of unemployment one ought

to take into account different participation rates in order to further standardize the

results. But it is clear that international unemployment performance in the leading

countries has noticeably weakened since the golden age days of Keynesian dominance

from the end of the war until the mid 1970’s.

The participation rate currently  in the UK is several percentage points below that

of Canada and even further below that of the US. For example, in 2000 the UK rate was

63.3, 65.9 in Canada and 67.2 % in the US.(Source R.Ehrenberg et al table 14.1,

p.484)Strictly speaking if we chose the US participation rate as the norm then

Canada’s unemployment would be worsened by an additional 1.3 percntage points and

that of the UK by an additional 3.9 points. However, it is arguable that lower

participation rates do not necessarily imply lower welfare. It might instead reflect more

available leisure and less necessity to work in the case of two member households.High

participation rates on the other hand usually reflect discouraged workers rejoining the

labour force but might also reflect lower wages and the necessity of more members of a

household to work to maintain the standard of living.

From Inflation to Disinflation and Deflation

Whereas unemployment in the post-war years right up until the OPEC price

shock in 1973/74 typically averaged 3-5 % the rates moved up sharply in the years that

followed.( See Table one )At the same time the inflation rates measured either by the

consumer price index or GDP deflator are very much lower than during the 1970s.  For

example, inflation in Canada ranged  between 9.1 and 15.3 % from 1973 to 1976,

averaging 11.2% for the four years, as measured by the GNE deflator.A similar sort of

sharp rise in inflation occurred in the US during the same period.It was this shock in

prices delivered by OPEC and the world wide boom in commodities prices that

established the anti-inflationary environment that has dominated the past 30 years in

macro-economic policy.

In fact, current evidence points toward the conclusion that we are on the edge of

deflation or falling prices, a situation that has prevailed in Japan for the past several

years, threatens parts of Western Europe and  was identified as  a threat in North

America several years ago. We are so used to fearing inflation and ignoring

unemployment that many leaders in the business and political world were blind -sided

when deflation suddenly appeared as a real prospect.

If one takes into account globalization, the internet, just in time production and

the flooding of markets by goods and services produced in  cheap wage countries like

India, Bangladesh, countries in South-east Asia  and China  disinflation and deflation

remains a very serious possibility in the years to come.China, like South Korea and Japan

before it,  will soon be exporting large numbers of cheaper automobiles to the North

American and European markets.This will put downward pressure on the price of

automobiles. Outsourcing of services and manufacturing has also had the same effect on

wages and prices.Only oil prices which are under the control of the OPEC oil cartel have

resisted this trend and it is not clear that this resistance can last over the longer term.

Business leaders like Jack Welch ,the president of General Electric,  raised the

spectre of deflation in 1998 in a letter to shareholders and the problem had been

discussed by Alan Greenspan of the Federal Reserve at the January 3rd  1998 American

Economics Association Annual Meeting . It had been a major topic of discussion at the

Federal Open Market Committee meeting the previous December. (See Alan Greenspan,

Problems of Price Measurement; Meeting of the Federal Open Market Committee,

Dec.16, 1997, Federal Reserve Board. cited in Chris Farrel, Deflation:What Happens

When Prices Fall

pp.20ffNew York: Harper Collins, 2004 See also Roger Bootle,  Death of Inflation,

London:Nicholas Brealey, 1996;and Harold Chorney “In Search of the non Walrasian

labour market model in the age of Globalization” paper presented to the Eastern

Economics Association, New York,Feb.23, 2001)

TABLE TWO:

UNEMPLOYMENT AND INFLATION AND BUDGET DEFICITS IN SELECTED OECD COUNTRIES OCTOBER 2004

Country                     Unemployment(U);   Inflation Rate(I)       Budget deficit (D ( -) as             % of GDP(+ surplus)

Country_________________U__________D____________I__________________________

Australia 5.6  (Sep)   2.5 Q2

Austria                          4.5  (Sep)   2.1(Sep)

Belgium                        13.2 (Sep)                                             2.0(Sep)

Britain                            4.7(Aug)              -3.4                         1.1(Sep)

Canada                          7.1  (Sep)            + 0.6                        1.9 (Sep)

Denmark                        6.3 (Aug)                                             1.1(Sep)

France                            9.9(Aug)               -4.1                         2.2(Sep)

Germany                         10.7(Sep)             -3.8                         1.8(Sep)

Italy                                 8.1(Apr.)              -2.4                         2.1(Sep)

Japan                               4.8(Aug)               -8.2                       -0.2(Aug)

Netherlands                      6.1(Sep)                                              1.0(Sep)

Spain                               11.0(Aug)                                             3.2(Sep)

Sweden                            5.8  (Sep)                                             0.6(Sep)

Switzerland                       3.7(Sep)                                               0.9Q2

United States                    5.5(Oct)               -4.6                          2.5(Sep)

Euro area                          9.0(Q2)                -2.8                          2.1(Sep)

__________________________________________________________________

Source: OECD, Economist, Oct.23, 2004 and Bloomberg News cited in

Financial Post, Nov.22, 2004.

The above data suggest a world in which inflation is very low but unemployment

far too high except in Britain,Austria, Japan and Switzerland. It hardly seems like a world

in which the operational policy doctrine ought to be one which argues that there is a

natural rate of unemployment below which a country experiences accelerating

inflation.Interestingly the country with the highest deficit to GDP, Japan also has the

second lowest unemployment and the lowest inflation, that is deflation.The Euro region

with a low deficit to GDP has one of the highest unemployment rates.The inflation rate

for the 8 entities for which budget deficits are given averages 1.7%.But unemployment

averages 7.5% for these same entities.France, Germany and the Euro area in general suffer

from very high unemployment. Not surprisingly the European Central Bank follows a

strict anti-inflation monetary restraint regime.

It is time for a radical shift in doctrine. It is time to abandon the non- accelerating

inflation rate of unemployment or the natural rate of unemployment and in its place

consider using the concept of the natural rate of inflation.

The Natural Rate of Inflation

The natural rate of inflation is that rate of inflation just slightly above or equal to the rate

that is consistent with a growing robust economy.It is also the rate of inflation below which

unemployment rises in an accelerating fashion.Just as Friedman’s concept built in

expectations of inflation, the natural rate of inflation is sensitive to expectations of future

rises in the rate of unemployment which can affect the consumption behaviour of

consumers, the animal spirits of investors and thereby overall aggregate effective demand.

In other words,  if  the central bank in its misguided attempt to suppress non existent or

dramatically weakened inflationary impulses tightens interest rates and constricts growth

so that the rate of inflation falls below this level it can provoke  a sharp rise in

unemployment which can become a chronic condition of the labour market in that

country. Far better for the central bank to recognize, that once inflationary impulses have

been so strongly weakened, that a steady but low rate of inflation is necessary to lubricate

investment decisions in times of uncertainty and provide for a steady growth in

employment to provide the necessary consumer confidence to sustain aggregate demand

to facilitate keeping unemployment low.(SeeA.J.Brown, “Accelerating Inflation and the

Growth of Productivity” in R.C.O.Matthews ed., Slower Growth in the Western World,

National Institute of Economic and Social Research, Policy Studies Institue&Royal

Institute of International Affairs, Joint Studies in Public Policy 6, London:Heinemann,

1982 ,pp45-60,  on how modest inflation can contribute to capital innovation and more

rapid growth)

So long as unemployment is significant and deficient aggregate demand is present,

monetary and fiscal impulses work largely, though not completely on the output side as

opposed to the price side of the price times output nexus. (P.O)Prices may rise as

unemployment is lowered  but the amount of price rise will be tolerable until we

approach more closely the point where further reductions in unemployment and

increases in growth trigger bottlenecks, disproportionalities and therefore sharp upward price

rises.This was a position that Keynes developed quite clearly in his chapter on prices in

the General Theory and also in his chapter on the fundamental equations in The Treatise

on Money (Vol.1, pp 133-150) Unfortunately Samuelson distorted his argument in a way that would

prove nearly fatal to the neo-classical Keynesian synthesis some 35 years

later.

Samuelson’s 45 degree supply curve or helping line directly contradicts Keynes’

curvilinear disproportional aggregate supply curve  which permits price rise well before

the point of full employment, as vector forces push first upon output but then gradually

push partly on output and partly on prices and then largely on prices and less on

output.(See J.M.Keynes, The General Theory of Employment,

Interest and Money, Collected Writings, Vol vii, 1936; Paul Samuelson, ”The Interaction

between the Multiplier Analysis and the Principle of Acceleration” in Review of Economic

Statistics xxi(May, 1939) Samuelson’s text Economics, An Introductory Analysis,New

York: McGraw-Hill,  1948, Harold Chorney, Keynes and the problem of inflation: The

Roots of the Return of Sound Finance, 1987 reprinted in The deficit papers,Montreal: 2002

. A French version of this paper is published in G.Dostaler&G.Boismenu,ed. La theorie

generale et le keynesianisme, Montreal: 1987. See also Harold Chorney, Revisiting Deficit

Hysteria in Labour/Le travail, 54(Fall 2004) pp.245-258.pp250-251 in  particular)

To better appreciate the concept of the natural rate of inflation  we need first to

examine the background to the development of the doctrine that it now should eclipse.

The Natural Rate of Unemployment and the Development of the NAIRU rate

The origins of the natural rate hypothesis can be traced back to the doctrines of the

natural rate of interest pioneered by economists like Knut Wicksell, Bohm Bawerk and

Hayek during the 1920s.(See Maurice Dobb, The theory of value and  distribution since

Adam Smith: Ideology and economic theory; Cambridge: Cambridge University press,

1973.R.W. Dimand, The Origins of the Keynesian Revolution, Aldershot: Edward Elgar,

1988; T.W.Hutcheson, The Politics and Philosophy of Economics:Marxists, Keynesians

and Austrians, New York: New York University press, 1984.Joseph Schumpeter,History

of Economic Analysis, David Laidler, Fabricating the Keynesian Revolution, Studies of the

Interwar Literature on Money, the Cycle and Unemployment, Cambridge: Cambridge

University press, 1999  &H.Chorney, The Theory of the Business Cycle in Keynes,Hayek

and Schumpeter, paper presented to the Society of Heterodox Economists, London,

2000.)

During the 1960s  inflation occured at rates that appeared to contradict the neo-

classical notion of Keynesian macro-economics that had come to dominate the discipline

under the hegemonic influence of the Samuelson-Hicks synthesis. In Canada, for example,

inflation had risen  from 0.9 percent in 1961 to 4.0% in 1968 to7.5 % in 1973 to 10.9 %

in 1974.Milton Friedman very cleverly articulated the doctrine of the natural rate of

unemployment. He did so most famously in his AEA presidential address, The role of

monetary policy in 1968(American economic review, March 1968 reproduced in Brian

Snowdon and Howard Vane, A Macroeconomics Reader,New York: Routledge, 199)

Friedman argued exactly as the classics had done many decades before that just as

there was an natural rate of interest that corresponded to a stable economic growth rate

and a normal rate of return of invested capital over time. This rate could not be lowered

without the peril of igniting a crisis of overproduction and overinvestment leading to a

crisis of inflation.(See Hayek and Bohm -Bawerk) Similarly there was a natural rate of

unemployment that the Walrasian system of general equilibrium ground out over the

medium and long haul.

As he put it ”  every attempt to keep  interest rates at a low level has forced the

monetary authority to engage in successively larger and larger open market purchases.

They explain  why  historically high and rising interest rates have been associated with

rapid growth in the quantity of money , as in Brazil or Chile or as in the United States in

recent years and why low and falling interest rates have been associated with slow growth

in the quantity of money, as in Switzerland now or  in the United States  from 1929-1933

“(Friedman, 1968)

He goes on to argue that low rates reflect a an initially tight monetary policy and

high rates paradoxically a loose monetary policy.In the end he argues a far better predictor

of  the tightness or looseness of monetary policy is “the rate of change of the quantity of

money. (p.169 in Snowdon and Vale) Friedman poses the question of why we cannot

simply target the rate of unemployment and be loose when the rate is above 3 %(How

extraordinary that he uses this most reasonble target by contemporary Keynesian

standards) and tight when the rate is below this rate. His answer is to appeal to Wicksell’s

concept of the natural rate of interest and point out the diference between short term

effects and long term consequences. The extremely close relationship between Wicksell’s

classical theory of the natural rate of interest is explicitly developed by Friedman in his

article.

Thus Friedman writes “Thanks to Wicksell, we are all acquainted with the concept

of a natural rate of interest and the possibility of discrepancy between the ‘natural’ and

the  ‘market’ rate. The preceding analysis of interest rates can be translated fairly directly

into Wicksellian terms. The monetary authority can make the market rate less than  the

natural rate only by inflation. It can make the market rate above the natural rate only by

deflation. We have added only one wrinkle to Wicksell – the Irving Fisher distinction

between the nominal and the real rate of interest.Let the monetary authority keep the

nominal rate below the natural rate by inflation. That in turn will raise the nominal natural

rate itself, once anticipations of inflation become widespread, thus requiring still more

rapid inflation to hold down the market rate above the initial ‘natural’ rate” (Friedman,

ibid, p.170)

Having developed his theory as a logical extension of classical theory Friedman

then reorients it to address the contemporary world of rising inflation in the late 1960s.

His goal which he largely achieved was to fracture the Keynesian consensus about the

wisdom of promoting lower unemployment through the judicious use of fiscal policy and

accomodating monetary policy. Friedman’s argument became the launch pad for a radical

restructuring of economic orthodoxy that has had a huge impact upon the conduct of

public policy and I would argue the actual rate of unemployment that is tolerated by the

leading G7 countries including the United States and Canada and Western Europe. Only

Japan and to a certain extent in recent years Great Britain have refused to sign on.

Any attempt to lower unemployment below this natural rate level through heroic

Keynesian stimulus would be doomed to fail as it would ignite inflation. In making this

argument Friedman was aided immeasurably by the reductionist Hicks-Samueson Hansen

model of the Keynes system that distorted Keynes original conception of  how the price

system would operate. Whereas Keynes had argued that stimulative fiscal and monetary

impulses would work their way through the system by acting largely on output and to a

lesser extent on prices so long as the economy was operating with unused capacity, then

gradually shifting so that vector forces would operate on the price side more than on the

output side and that the factors were often disproportional and non homogeneous(see ch.

21 of Keynes’s General Theory), Samueslon had unfortunately abandoned this more

sophisticated and realistic model for one which claimed that aggregate demand inflation

was the most powerful likely outcome. His 45 degree supply curve ( which he first

introduced in an article on the multiplier and accelerator in 1939 and later became the

standard orthodoxy through his best seller text book) represented the  aggregate supply

curve which implied linearity and inflation only once full employment output had been

reached. In such circumstances Keynesian theory was an easy target for Friedman’s

monetarist counter-revolution when the time came during the inflationary 60’s and

stagflationary 70s.

Friedman also posed his argument in the context of the policy framework that

dominated in the 1960s. This framework was expressed through the Phillips curve

tradeoff that argued that unemployment traded off with inflation.  Phillips had compared

wage rate changes with rates of unemployment over more than a century. He concluded

that there was a tradeoff expresed by the Phillips curve.- a partly convex curve to the

origin that traded increases in the unemployment rate on the x axis with decreases in

inflation on the y axis. Over time this Phillips curve had drifted out from the origin so that

the terms of the trade-off had seemed to worsen.

Friedman’s position was that the trade-off was illusionary once money illusion

had been banished. In other words once people built expectations about future price

inflation into their wage choices and therefore employment decisions, the trade offs that

the Keynesians had theorized about simply disappeared. There was, in fact, a vertical

Phillips curve where any rate of inflation was compatible with the natural rate of

unemployment.Friedman called this curve the expectations augmented Phillips curve. It

was vertical at the point of full employment, the so called natural rate below which the

economy would experience accelerating inflation.

In practice Friedman’s vertical Phillips curve was a utopian concept that could

not be sustained by the evidence in the real,  as opposed to imaginary world of abstract

economics. Instead of a perfectly vertical curve based on the NAIRU rate of

unemployment Friedman and his followers  accepted that, at least, in the short run there

might be some rise in the rate of unemployment as inflation was wrung out of the

system.But as this was accomplished the curve would revert to the nearly vertical

position  in the long run.

Thus the distinction was born between the short run and long run Phillips curve.

Whatever unemployment the policy of monetarism caused would be acceptable because it

would be a short lived phenomenon.Of course reality was much more unforgiving.

Unemployment rose sharply as monetary policy was tightened and unemployment was

very slow to fall back to its original position. In fact, in some cases it never did as millions

of people in western countries lost their jobs never to return to the labour market again.

Economists began to notice that unemployment scarred people for a long time and

prevented some of them from finding work again even when the recession that Friedman

inspired policies had induced was over. The notion of hysterisis was developed to speak

to this problem.

I also believe that Friedman’s curves need to be rethought in a fundamental way.

First it seems to me that the experience of countries like Canada and the UK when

monetarist policies were applied demonstrated that while the short run experience might

initially only raise the unemployment rate a few percentage points, at a certain point

people would be frightened by the rise in unemployment and these widespread

expectations that unemployment was growing would result in an even sharper rise in the

rate of uneployment that would accompany the drop in inflation. Hence the long term

Phillips curve at the point of deflection where expectations would turn pessimistic would

swing out to the right.

Initially in the short run curve inflation is brought down and

unemployment rises in response to tight monetary policy. Then at point B where the

unemployment rate begins to rise sharply  the long run

curve comes into play and unemployment rises much faster than inflation falls.

Eventually the economy reaches a point of high unemployment and low inflation..The

victory over inflation has been achieved but at a very high cost(not Pareto compensated

either) , We now have a situation of chronic high unemployment that may last several

years coupled with low inflation.

It was only a matter of time for the natural rate argument to become dogma in the form

of the NAIRU rate and the classical world of rational expectations to treat chronic

unemployment under the rubric of voluntary unemployment and inefficient job search.

It is important to understand how contemporary neo-classicals view

unemployment. Armed with the notion of the natural rate of unemployment and the

NAIRU rate they have come to  see labour markets operating as fairly efficient market

clearing mechanisms where involuntary unemployment is limited . The most extreme

rational expectations neo-classicals presume that labour markets, like  other markets,  are

super efficient allocators of tastes and preferences and wants. In addition the labour

market is a perfect information transfer system. If some one needs a job,  efficient job

search will match the job seeker with the job at the market clearing wage. Walrasian

general equilibrium will hold with little difficulty. Some frictions may be present but they

will be overcome in fairly short order and labour will be efficiently allocated. Any

remaining unemployment occurs because these workers do not search efficiently for jobs

because they have set their reservation wage too high either because of union pressure or

the presence of too generous unemployment insurance benefits; or perhaps their

intellectual capital in the form of education and training make them unsuitable for available

jobs; or perhaps they simply prefer leisure to work and in effect have dropped out of the

labour market.Or perhaps other rigidities prevent the labour market from finding the right

market wage to clear the market of all those seeking work. Whatever the reason

involuntary demand deficient Keynesian unemployment is banished from their model.

The model of how the labour market works is based upon stocks and flows. There

is a stock of employed people, a stock of unemployed people.One measures job

separations that is quits or lay-offs and also job findings that is hiring of job seekers from

the pool of unemployed workers and the flow of new entrants.

The unemployment rate is the outcome of these contrary flows and can be

measured precisely.

Overturning Friedman

My concept of the natural rate of  inflation turns Friedman on his head. Imagine a

rate of inflation compatible with steady growth and low unemployment. At this rate of

inflation much of the curve is close to horizontal implying that any rate of unemployment

is compatible with the natural rate of inflation. Of course, over the longer run as we push

the unemployment rate down towards 2-3 % we will begin to experience bottleneck

inflation and also wage and profit push pressures. In such circumstances inflation may

begin to rise above the natural rate if we use aggregate demand to lower the rate of

unemployment. The curve thus may not strike the Y axis at the natural rate but turn up

before contacting the y axis but at a much less steep angle than Friedman’s.Hence, just as

in the Friedman case there is a distinction between the short and long run.

The  behaviour of this natural rate of inflation is quite straightforward..

Let P* = the rate of inflation. h= expectations about the future. hU -expectations of

future unemployment. hP* =expectations of future inflation. W* = the rate of wage

increase. U= the rate of unemployment.P*n =the natural rate of inflation. U*n = the

natural rate of unemployment; s*=rate of unionization; o*degree of oligopoly

concentration, a=investor animal spirits.t=state of technological innovation, g =degree of

globalization.

Then P*n = f( hU, hP*,w*, s*,o*,a,t, g) the greater s*,o* hP* the sharper the

angle of deflection upwards( measured between the vertical Phillips curve and the

deflected line from point B) and the higher the rate of inflation that deflection occurs at;

hU,g and ‘a’ have the opposite effect.The greater they are the smaller the angle of

deflection. t will work in both directions. On the one hand innovation lowers the price of

goods and improves efficiency of output and increases quality. On the other hand,

innovation may also raise the rate of unemployment , at least in the short run by

substituting capital for labour.

The original Friedman natural rate was modeled as an expectations augmented

Phillips curve which rose vertically from the x axis at the natural rate of unemployment.

This then became the NAIRU rate, the non accelerating inflation rate of unemployment.

It was modeled as follows: Assume initially that inflationary expectations are zero. Then

(dP/P)* = 0 .where (dP/P)* is the expected rate of inflation. Assume that the economy is

now at U’. Then U=U’ at point A on the x axis and the Phillips curve which cuts the axis

at this point.Thus U=U’ and dP/P = (dP/P)* Unemployment is at its natural value and

expected and actual rates of inflation are equal. Now assume that an activist government

seeks to trade-off some inflation for a reduction in unemployment. It chooses to move to

position B. Initially, according to Friedman unemployment

will drop to U1 corresponding to point B on the Phillips curve.The unemployment

inflation pairing U1,(dP/P)1, according to Friedman is unstable. Because at B  the

expected rate of inflation (dP/P)* will no longer remain at zero but instead climb.(dP/P)*

when (dP/P)1>(dP/P)*. More specifically d/dt(dP/P)= z((dP/P)-(dP/P)*. Expectations are

adaptive and the parameter z measures the speed with which expectations adapt to past

errors in predicting the rate of inflation. Any attempt , according to Friedman to hold the

unemployment level below the point where expected inflation is zero will result in an

actual rate of inflation that exceeds the expected rate.(See J.A.Trevithick, Involuntary

Unemployment :Macroeconomics from a Keynesian Perspective Harvester

Wheatsheaf::New York, 1992,  pp,132ff)

Friedman was arguing that the rate of inflation could be decomposed into two

portions. The first an expectational component had the form (dP/P)* and second an

excess demand component measured by the original Phillips -Lipsey model by f(U) The

natural unemployment rate was that unique rate U’ where the excess demand component

f(U) was zero. Any rate of  inflation, positive or 0 or negative is compatible for Friedman

with the natural rate U’ since the only condition for equilibrium is that inflationary

expectations be fully realized so that dP/P=(dP/P)* (Trevithick, p.134)

Trevithick suggests that one can also regard the accelerationist hypothesis as

follows.

(dP/P) – (dP/P)*=f(U) and because of  the assumption of adaptive expectations then

d/dt(dP/P)*=zf(U).As Trevithick points whatever trade-off that Friedman accepts is the

trade-of between the rate of unemployment and the “rate of change of the actual and

expected rates of inflation..” This is a temporary and not permanent trade-off which

comes about from unanticipated inflation.(Friedman, ” The role of monetary

policy “, American Economic Review, vol.58, 1.)

With this background in mind let us construct a parallel set of reasoning for the

natural rate of inflation. Assume that the economy is at a rate of inflation compatible with

stable unemployment and growth. There are no differences between the expected rate of

inflation and the actual rate. If anything differences may be working in the opposite

direction. That is the actual rate is below the expected rate. So that dP/P< (dP/P)*.

Similarly dU/U = (DU/U)* that is the actual rate of unemployment is equal to the

expected rate. Any attempt to move the inflation rate below this natural rate will lead to

the actual rate of unemployment to exceed the expected rate and unemployment will rise.

That is,  dU/U > (du/U)*. Furthermore once we are at position B1 on the y axis and

position B on the curve A1BA3, any rate of unemployment will be compatible with the

natural rate of inflation provided that the expected rate and the actual rate remain the

same. Position B is also a point on the lines DBF and DBB1.At some point , it will vary

with each economy, as we attempt to lower unemployment through both monetary and

fiscal policy prices may well rise and the curve will turn upwards before touching the Y

axis, as in DBF as we move from unemployment U’ to U1 and U2.

We can better understand the analysis if we imagine that we begin with a

downward descending Phillips curve but not a vertical one.In other words, the line

A1BA3 as opposed to the vertical line A2BA, the original expectations

augmented Phillips Curve which Milton Friedman postulated in 1968. Rather one that

many economists have suggested is the short run Phillips curve. As the rate of inflation

drops and the rate of unemployment rises we approach position B on the diagram. At

this point expectations of unemployment begin to rise as the rate of unemployment

increases in response to contractionary monetary and fiscal policy. Below this point

which I am arguing corresponds to the natural rate of inflation as measured on the Y axis,

unemployment rises sharply as the curve deflects and moves dramatically out to the right.

This then results in the actual rate of unemployment exceeding the expected rate which

results in a deep prolonged recession. Inflationary expectations at this point vanish and

converge around the actual rate of inflation. At this point we have reached the zone of the

natural rate of inflation.

Any attempt to push inflation below this level will result in a prolonged period of

high unemployment that is largely due to deficient agregate demand and depressed

expectations. Instead,  if the monetary authorities avoid squeezing interest rates any

higher but reduce them and indeed accommodate some fiscal stimulus we should experience

a considerable reduction in the rate of unemployment without significant price rise until

our level of unemployment approaches U’ at point B. Then once B is reached or possibly

even surpassed prices will begin to rise but more gently, that is, less rapidly than the

original curve suggests.The  curve DBB1 or DBF is the short run curve with DBB1 being

the horizontal equivalent to the vertical expectations augmented Phillips curve.A more

realistic curve for the longer run is DBF or perhaps DBE.

Unemployment can be pushed below the natural rate of unemployment to either

U1 at inflation rate H1 or possibly even lower to U2 at inflation rate H2.

More formally BB1 is P*n, the natural rate of inflation where dp/P=(dP/P)* and

dU/U>(DU/U)*. Then if so,  U rises to approach Uc corresponding to point C on the line

BC as the central bank continues its contractionary policy to force inflation below B1. If

the central bank then relaxes its policy and fiscal policy is supportive then U is reduced

gradually  toward B.

In such circumstances BCC1 may be unstable  and the curve may rise to

DB1.The area bounded by BDC is the zone of the natural rate of inflation.The area BB1F

is the zone of low unemployment and moderately low  inflation. If we push

unemployment below OA then price rise will occur but the trade-off will be much more

favourable and long lasting than in the past. It is unlikely that the path B1BD can last

beyond the short run. Instead DBF is more likely to prevail. This implies some price rise

but much more slowly and less drastically than the original curve A1BA3.

As we move back toward B the gap between dU/U – (dU/U)* decreases and

approaches zero. It reaches  zero at either B or perhaps G a point just to the left of B.

Finally d/dt(dU/U)*t+1 > d/dt(dU/U)*t.

Conclusion:

It could be the case that the natural rate of inflation sets the lower boundary to the

upper boundary for the natural rate of unemployment and that the economy swings

between these two positions over a long period of time and policy paridgmatic shift. But

whatever the case,  the natural rate of inflation shows promise as a concept whose time

has arrived.An understanding of it would permit central banks and governments to

address the excessive unemployment rates that now plague most of the western world.

They would need  to use the levers of monetary and fiscal policy to facilitate significant

reductions in unemployment and restore fuller employment. This means both keeping real

interest rates low and using fiscal stimulus in the form of  reduced surpluses or actual

planned ex ante deficits targeted at low and moderate income people and the educational,

physical and health care infrastructure. Only when the rates of unemployment have

dropped to much lower levels would it be time to work to restore balanced

budgets.through appropriate taxation measures.

Appendix

TABLE ONE

UNEMPLOYMENT IN CANADA, THE UNITED STATES

and GREAT BRITAIN

1930-2001

_____________________________________________________

CANADA     UNITED STATES BRITAIN

1930        9.1 %        8.7 %                 11.1 %

1931       11.6         15.2                   14.8

1932       17.6         22.3                   15.3

1933       19.3         20.5                   13.9

1934       14.5         15.9                   11.7

1935       14.2         14.2                   10.8

1936       12.8          9.8                    9.2

1937        9.1          9.1                    7.7

1938       11.4         12.4                    9.2

1939       11.4         17.2                    5.8

1940        9.2         14.6                     3.3

1941        4.4          9.9                     1.2

1942        3.0          4.7                     0.5

1943        1.7          1.9                     0.4

1944        1.4        1.2                     0.4

1945        1.6          1.9                     0.5

1946        2.6          3.9                     1.9

1947        1.9          3.9                     1.4

1948        1.6          3.8                     1.3

1949        2.0          5.9                     1.2

Unemployment

Canada           US                Britain

1950        2.0          5.3                     1.5

1951        1.5          3.3                     1.2

1952        2.0          3.0                     2.0

1953        3.0          2.9                     1.6

1954        4.6          5.5                     1.3

1955        4.4          4.4                     1.1

1956        3.4          4.1                     1.2

1957        4.6          4.3                     1.4

1958        7.0          6.8                     2.1

1959        6.0          5.5                     2.2

1960        7.0 %        5.5 %                 2.2 %

1961        7.1          6.7                   2.0

1962        5.9          5.5                   2.8

1963        5.5          5.7                   3.4

1964        4.7          5.2                   2.5

1965        3.9          4.5                   2.2

1966        3.4          3.8                   2.3

1967        3.8          3.8                   3.4

1968        4.5          3.6                   3.6

1969        4.4          3.5                   3.0

1970        5.7          4.9                     3.1

1971        6.2          5.9                     3.7

1972        6.2          5.6                     4.7

1973        5.5          4.9                     2.9

1974        5.3          5.6                     2.9

1975        6.9          8.5                     4.1

1976        7.1          7.7                     5.5

1977        8.1          7.1                     6.2

1978        8.3          6.1                     6.1

1979        7.4          5.8                     5.8

1980        7.5          7.1                     7.3

1981        7.5          7.6                    10.4

1982       11.0          9.5                    12.0

1983       11.8          9.5                    13.2

1984       11.2          7.5                    13.6

1985       10.5          7.2                    11.8

1986        9.5          7.0                    11.8

1987        8.8          6.2                    10.6

1988        7.8          5.5                     8.4

1989        7.5          5.3                     6.3

1990        8.1          5.5                     6.0

1991       10.3          6.7                     8.4

1992       10.5          7.5                     10.3

1993       10.8          7.0                     10.7

1994       10.6          6.1                     9.8

1995   9.5  5.6  8.8

1996        9.6          5.4                     8.2

1997        9.1          4.9                     7.1

1998        8.3          4.6                     6.3

1999        7.6          4.2                     6.1

2000        6.8          4.0                     5.5

2001                 7.2                     4.7                                           5.1

2002                 7.7                     5.8                                           5.2

2003                 7.6                    6.0                                           5.0

2004 (Oct.)       7.1                    5.5                                           4.3

Source : Statistics Canada; US Bureau of Labour Statistics; OECD; Ontario Economic

outlook and Fiscal Review, 2004.,B.R.Mitchell&P.Deane, Abstract of British Historical

Statistics,.