SOME REFLECTIONS ON MACROECONOMIC POLICY: WHAT NEEDS TO BE DONE TO SUSTAIN GROWTH AND ACHIEVE A FULLY-EMPLOYED ECONOMY B Y M A R I O S E C C A R E C C I A ( U N I V E R S I T Y O F O T T A W A )
WHAT WAS THE RECORD OF ACHIEVEMENT? Figure 1 traces the unemployment rate in Canada from the golden age era of 1946-1973 and post-golden age period since 1973. This pattern of unemployment is not specific to Canada but it can be found in practically all Western industrialized countries. One notices a clear break during the 1970s, thereby suggesting that something happened during that post-1973 era that set these economies on a different trajectory. In fact, in Canada we had achieved the lowest rate of unemployment in 1966 at 3.6 percent; but with a rate gravitating around a 4.6 percent average (from 1946-1973). During that era when I began studying economics in the late 1960s, most economists thought that, given the industrial structure and regional diversity of the Canadian economy, a full employment rate of unemployment of around 3 percent was a feasible goal (e.g. Economic Council of Canada 1965).
FIGURE 1: UNEMPLOYMENT RATE IN CANADA, ANNUAL, 1946-2011 14 12 10 Percent 8 6 4 2 50 55 60 65 70 75 80 85 90 95 00 05 10 Unemployment Rate Average Unemployment Rate 1974-2011 Average Unemployment Rate 1946-1973
EXPLANATIONS THAT JUSTIFIED POLICIES THAT ACCEPTED RISING UNEMPLOYMENT With the rise of the unemployment rate (with a post-1973 average of 8.4 percent) and with the concomitant rise in choice-theoretic theories of unemployment that were largely apologetic and that argued that this rise was the result of natural tendencies pertaining to the supply-side behaviour of the unemployed themselves, the concept of full employment seemed to have withered away. The achievable unemployment rate now termed the natural rate of unemployment (NRU) or the non-accelerating inflation rate of unemployment (NAIRU) rose so that, by the late 1980s, the NAIRU now had reached ever higher levels in the range of 7.5-8.5 percent (as estimated at the Bank of Canada in the late 1980s). But were these supply-side explanations meaningful or what is it simply because of lack of demand pertaining to policy choice over the last three decades?
HARRODIAN NATURAL AND WARRANTED GROWTH RATES: FROM A GOLDEN AGE TO A LEADEN AGE There is a famous equation from Harrod (1939) defining the rate of growth of potential output that neoclassical economists such as Solow (1956) had subsequently interpreted as the rate around which the actual economy tends: G* = η + λ where η is the labour force growth and λ is the growth in productivity. What Harrod had said was quite simply. If we think of a fully-employed economy, potential output must, by definition, grow at the rate at which the labour force is increasing, η, plus the rate at which productivity, λ, is rising. The actual G may deviate from G*, depending on entrepreneurial animal spirits that determine the warranted path of capital accumulation and growth.
FRAMING EMPLOYMENT POLICY IN A HARRODIAN FRAMEWORK I would like to use this same Harrod equation of the natural rate of growth to frame the policy question of achieving full employment. Following Robinson (1962), an economy in which G=G* must be one following a Golden Age full-employment growth path. Instead an economy in which G < G* is in a so-called Leaden Age in which unemployment continues to rise in the longterm. In a Golden Age economy, there is sufficient growth in output and, concomitantly in employment, to absorb the growth in the labour force, η, but also to absorb the amount of technologically-displaced labour that the economy is shedding as a result of technical change, λ.
REINTERPRETING THE HARROD NATURAL RATE OF GROWTH EQUATION For the economy to maintain a constant unemployment path, there needs to be sufficient growth in aggregate demand to prevent the unemployment rate from rising over time. Hence if we look at the labour force growth, it was 2.4 percent during the 1947-1973 period but it fell to 1.8 percent growth for the 1974-2011 period, as shown in Figure 2 below. On the other hand, productivity growth grew by an average of 4.0 percent during the 1947-1973 period, but only by 1.4 percent average from 1974-2011 period in Canada! This would suggest that one needed over 6 percent growth in output to maintain a constant unemployment rate during the pre-1973 period in Canada; while it only needed about 3 percent growth to do the same for the post-1973 period and today. Yet, when we look at the unemployment picture, shown in Figure 1, we saw that the unemployment rate rose a great deal, especially during the 1980s and part of the 1990s, because of the obvious anemic growth of the period.
FIGURE 2: LABOUR FORCE GROWTH IN CANADA, 1947-2011, AND SUB-PERIODS 5 4 3 Percent 2 1 0-1 50 55 60 65 70 75 80 85 90 95 00 05 10 Labour Force Growth Average of Labour Force Growth, 1974-2011 Average of Labour Force Growth, 1947-1973
FIGURE 3: PRODUCTIVITY GROWTH IN TOTAL BUSINESS SECTOR IN CANADA, 1947-2011, AND SUB-PERIODS 10 8 6 Percent 4 2 0-2 50 55 60 65 70 75 80 85 90 95 00 05 10 Average Labour Productivity Growth, Total Business Sector Average for 1974-2011 Average for 1947-1973
WHAT DOES THIS SAY FOR EFFECTIVE EMPLOYMENT POLICY? What these stylized facts highlight is that it is much easier to get the unemployment rate down via expansionary macroeconomic policy today because of lower labour force growth and lower productivity growth than during the pre-1973 Golden Age ; yet we are living in an era of stagnation consistent with a Leaden Age. For instance, in the case of fiscal policy, an extra $x of government spending today will be a lot more effective in getting the unemployment rate down than the $x that were needed during the Golden Age to achieve the same goal. Or, to state this differently, to maintain the unemployment rate constant, one requires a lot less public spending today than during the pre-1973 period. Yet, unemployment has continued its long-term rise! We can certainly conclude that it is not primarily the supply-side that was moving the system to generate this rising unemployment phenomenon of the Leaden Age. It s primarily demand-side problem, namely a choice on the part of the current and past generations of policymakers that desired to keep the economy at such a level, supposedly to keep a lid on inflation.
THE MACRO POLICY MIX WAS THE CULPRIT The macro policy mix for the post-1973 period was one of very tight monetary policy of high real interest rates, which had destabilized the public finances that triggered high budget deficits, especially during the two deep recessions that ensued (in 1981-82 and 1990-91 in Canada) and began to scare policy makers to combat budget deficits. However, since the mid-1990s the policy mix changed dramatically in favour of looser monetary policy and very tight fiscal policy until the financial crisis of 2008. In the case of fiscal policy, as shown in Figure 4, we could see the change that took place starting in the mid-1990s, which in the case of Canada was actually able to achieve a string of public-sector surpluses until the financial crisis.
FIGURE 4: GENERAL GOVERNMENT BALANCES AS A PERCENTAGE OF GDP, CANADA AND G-7 AVERAGE 4 2 0 Percent -2-4 -6-8 -10 1970 1975 1980 1985 1990 1995 2000 2005 2010 Canada General Government Financial Balances as % of GDP G-7 Average General Government Financial Balances as % of GDP Source: OECD Economic Outlook, No. 91 (May 2012); Department of Finance Canada, Fiscal Reference Tables (October 2012).
EFFECTS OF THIS CHANGE IN FISCAL POLICY POSITION The fiscal policy change entailed drastic cuts in discretionary government expenditures, as well as important destabilizing institutional changes: 1. Cuts in discretionary federal program spending, such as federal transfers to other levels, as well as continued pressure to bring down public investment expenditures to very low levels. 2. Tinkering with automatic stabilizers, especially the employment insurance (EI) program.
FIGURE 5: EVOLUTION OF FEDERAL TRANSFERS TO OTHER LEVELS OF GOVERNMENT AS A PERCENTAGE OF GDP AROUND ITS H-P TREND (QUARTERLY OBSERVATIONS), 1961-2011 5.5 5.0 4.5 Percent 4.0 3.5 3.0 2.5 65 70 75 80 85 90 95 00 05 10 Source: Statistics Canada, CANSIM series V498086 and V498377.
FIGURE 6: EVOLUTION OF GOVERNMENT FIXED CAPITAL FORMATION AS A PERCENTAGE OF GDP, CANADA, 1990-2012 4.4 4.0 3.6 Percent 3.2 2.8 2.4 2.0 90 92 94 96 98 00 02 04 06 08 10 12 Share of Public Investment as a Percentage of GDP Source: Statistics Canada, CANSIM series V498086 and V4980093.
FIGURE 7: ANNUAL FEDERAL EMPLOYMENT INSURANCE ACCOUNT BALANCES AS A PERCENTAGE OF FEDERAL REVENUES, 1983/84 2011/12 6 4 Percent 2 0-2 -4 84 86 88 90 92 94 96 98 00 02 04 06 08 10 EI Balance Source: Department of Finance Canada, Fiscal Reference Tables (October 2012).
FIGURE 8: INDICATOR OF AUTOMATIC STABILIZER COMPONENT OF BUDGETARY BALANCES AS A PERCENTAGE OF GDP, ALL LEVELS OF GOVERNMENT, 1963-2011 4 3 2 1 Percent 0-1 -2-3 -4 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Automatic Stabilizer Component Source: Statistics Canada, CANSIM Table 380-0007.
THE FISCALIST PERSPECTIVE AND THE POLICY PROUETTE FROM 2008-2010 The policy changes that took place immediately following the financial crisis were critically important and were not much different to the policy reactions during the 1930s: (1) the bringing of interest rates to their lower bounds; (2) fiscal stimulus, primarily through public investment, and extending of transfers to the unemployed. Problem: in accordance with the New Consensus in macroeconomics, these stimulus programs were seen as temporary measures that needed to be reversed as soon as the economy began to show signs of growth (the green shoots ), because of fears of destabilizing inflation, and because of their supposed fiscal unsustainability. It is because of these fears these fiscal stimulus measures were quickly reversed starting in 2010 after the G-20 meeting in Toronto.
KEYNES S RADICAL VISION SHOULD BE TODAY S POLICY VISION TO ACHIEVE FULL EMPLOYMENT There are two policy proposals in the General Theory which are both long-term visions of policy: (1) the so-called euthanasia of the rentier, and (2) the socialization of investment. The first entailed simply the abandonment of Wicksellian policy of stabilizing real interest rates and therefore rentier income over time, and the replacement with the pegging of nominal interest rates at their lower bound. (This latter policy was largely central bank policy during the financial crisis but this ought to be permanently in place). The second was the socialization of investment which (as I have argued elsewhere (Seccareccia 2011-2)) was essentially a long-term growth policy based on the significance of a sustained public investment policy over time that would provide a makeweight around which private investment would eventually gravitate. Both of these policies were implemented in a half-baked manner immediately following the financial crisis (in 2009-10) but then were quickly abandoned. These Keynesian policies ought to be of a longterm nature to achieve high growth and veritable full employment.
THE RESTORATION OF STRONG AUTOMATIC STABILIZERS While these two Keynesian policy measures would ensure that the economy would be achieving potential output growth in the long term, short-term fluctuations due to short-term changes in private sector spending could still move the economy away from its potential path cyclically. For this reason, there would still be the need to have in place strong automatic stabilizers that would ensure effective full employment whenever private sector demand is insufficient. This is why Minsky s support of a strong form of Lerner s functional finance via some form of ELR proposal would also be essential to ensure that the economy would remain tightly on its full-employment growth path and prevent anyone from being involuntarily unemployed.