Economic Reform in New Zealand 1998 PDF

* Economic Reform in New Zealand: Comment Dr Paul Dalziel Reader in Economics Departm...

* Economic Reform in New Zealand: Comment Dr Paul Dalziel Reader in Economics Department of Economics and Marketing P.O. Box 84, Lincoln University, Canterbury, NEW ZEALAND E-mail Address: [email protected] Abstrakt Evans, Grimes and Wilkinson's (1996) essay on New Zealand's economic reforms claims that real per capita income has risen above its pre-reform trend, overseas debt has fallen, and the number of households experiencing poverty is no greater than before the reform program. This communication documents that the data producing these claims are not reliable. Instead, there is still no clear-cut measured improvement in New Zealand's growth performance, overseas debt was significantly higher in 1995 than in 1984, and substantial income sacrifices between 1988 and 1993 doubled the number of households in poverty. Keywords: Economic reform; Structural Adjustment; New Zealand. JEL Classification: E65, O56 (1 April, 1998) * This communication is based on a longer paper presented to the Annual Conference of the New Zealand Association of Economists in August 1997. I am grateful to Arthur Grimes, Steve Edwards, David Rae and Ashley Lienert for answering questions I posed on the data sources for the original essay by Evans et al. (1996), to Bryce Wilkinson for his comments on my presentation at the NZAE conference, and to Simon Chapple, Srikanta Chatterjee, Julia Crouch, Ross Cullen, Brian Easton, Viv Hall, Tim Hazledine, Jane Higgins, Stephen Jenkins, Jane Kelsey, Ralph Lattimore, Claudio Michelini, Caroline Saunders and Bob Stephens for discussions on earlier drafts. 1  ECONOMIC REFORM in NEW ZEALAND: COMMENT I. Introduction The December 1996 edition of the Journal of Economic Literature contains an important essay by Lew Evans, Arthur Grimes and Bryce Wilkinson with David Teece entitled 'Economic Reform in New Zealand 1984-95: The Pursuit of Efficiency'. That essay, hereafter referred to as EGW, summarizes and evaluates key aspects of New Zealand's reform program between 1984 and 1995 before concluding: 'After decades of policy errors and investment blunders, New Zealand appears to have finally diagnosed its predicament appropriately and is on a trajectory to maintain its economy as a consistent high performer among the OECD' (p. 1895). This unqualified conclusion does not represent a consensus view within New Zealand. Two special sessions at the 1996 and 1997 annual meetings of the New Zealand Association of Economists, for example, contained strong debate about the success of the reform program, focusing on the lack of clearly-measurable and sustained improvements in ultimate goals such as higher growth, improved productivity, lower unemployment and reduced poverty. The fundamental issue has been expressed by Brian Silverstone, Alan Bollard and Ralph Lattimore in their introduction to the most thorough set of essays evaluating the reform program to date. After observing that the reforms were unusually comprehensive, stemmed from a consistent theoretical framework, and were implemented within a relatively short period, these three senior economists go on to observe (1996, p. 19): If ever the impacts of economic reforms were going to be identified and studied in a developed economy, New Zealand provides that opportunity. Even such a 'big bang' shock on the scale that New Zealand experienced, however, has proved surprisingly difficult to measure. The reforms were clearly major, yet a number of performance variables did not register the expected impact. In contrast, EGW support their conclusion with claims that real per capita income in 1995 was 6.8 percent higher than the pre-reform average growth rate would have predicted (p. 1872), that total overseas debt was reduced by 17 percentage points of GDP between 1984 and 1995 (p. 1872), and that the number of households in poverty was no greater (and may have been 2.9 percentage points lower) in 1993 than in 1984 (p. 1882, n. 23). The purpose of this comment is to demonstrate that these figures do not stand close scrutiny. II. Poverty in New Zealand It is convenient to consider the three claims in reverse order. EGW acknowledge 'a widening dispersion of household incomes' (p. 1882), but then present the following reference in note 23: Bob Stephens, Paul Frater, and Charles Waldegrave (1995) provide evidence that, on a relative income basis, poverty fell, or remained unchanged, over the full 1984-93 period. When the poverty line was defined to be 60 (50) percent of median income the percentage of households in poverty changed from 13.7 (4.3) to 10.8 (4.3) over the period. They also discussed poverty from an absolute income standpoint. This represents a highly selective extract from the Stephens, Frater and Waldegrave article (hereafter referred to as SFW), and does not reflect the wider New Zealand literature about poverty during the reforms. SFW's full conclusion was the following (p. 109): If [the poverty standard] was adjusted in line with movements in median equivalent household disposable income, implying a relative view of poverty, then the combined 2  impact of a sluggish economy, economic restructuring and a 10% cut in social security benefit levels impacting on the critical income group resulted in a 17.6% drop in median income, but no measured increase in poverty. If an absolute poverty standard is used, the incidence of poverty before adjusting for housing costs more than doubled, from 4.3% in 1984 to 10.8% of households in 1993. EGW report the 'no change in relative poverty' result, but not the rise in absolute poverty. Yet this is an unusual result that is likely to be of wide interest, since absolute poverty standards typically produce falling poverty measures as per capita GDP grows, even if relative poverty increases. New Zealand did not conform to this pattern because the fall in median income of 17.6 percent meant that the relative poverty standard used in the SFW study (50% or 60% of median income) fell by the same percentage, as Easton (1995, p. 200) explains: To understand why this has happened, consider the following thought experiment, not unrelated to what actually occurred over the period. Take an income distribution and calculate the headcount poverty based on some proportion of the median as the measure. Now suppose the government taxes those on middle incomes, and transfers the revenue gains to the rich in lower taxes. The mean income will not change, but the median (which is based solely on the distribution of middle incomes) will fall. As a result, the [median datum line] will fall, and so the numbers in poverty will fall. SFW's conclusion that absolute poverty rose during the reforms to 1993, particularly after social welfare cutbacks in April 1991, has been supported in empirical studies by Krishnan (1995) and Easton (1996), leading to what Kelsey (1997, Chapter 11) has termed 'a social deficit' in the New Zealand economy. III. New Zealand's Overseas Debt EGW (p. 1872) argue that New Zealand's post-reform growth is likely to be more sustainable than previously, on the basis of 'a major reduction in the overseas debt to GDP ratio (78 percent in June 1995, compared with 95 percent in June 1984...).' Official data on overseas debt are notoriously weak in New Zealand-Statistics New Zealand's current series on total overseas debt began only in December 1992, and the previous series (which did not include domestically issued securities held by overseas residents) began only in September 1989. Prior to this, long-term overseas debt statistics are available as at the end of March from 1983 to 1986 and on a quarterly basis thereafter. Concern that this poor statistical base was discouraging overseas lenders led to a study of New Zealand's overseas debt trends by the New Zealand Institute of Economic Research (NZIER), undertaken by Pat Colgate and Joselyn Stroombergen (1993). In contrast to EGW's claim, the NZIER series has total overseas debt (excluding domestically issued securities) at 49.2 percent in March 1984, rising to 68.3 percent in March 1995. The data used by EGW were supplied by the Economics Division of one of New Zealand's leading commercial banks. This series includes domestically issued securities held by overseas residents, but this does not explain the divergence from the NZIER data. In March 1984, domestic interest rates were subject to regulatory ceilings, the exchange rate was fixed at a value that was becoming more and more overvalued (leading to a 20 percent devaluation in July that year), and there were restrictions on the ability of overseas-owned companies to borrow on the New Zealand capital market (removed in November 1984). Hence, the amount of domestically issued securities held by overseas residents could not have been more than in the deregulated environment of the 1990s, and was probably close to zero. Instead, the divergence is caused by the way in which the respective series estimate short-term debt before 1989. In the NZIER study, the authors adopted a Reserve Bank one-off survey of short-term debt in 1983 as a base, and then used Balance of Payments data to construct a series of implied short-term 3 debt movements to finance ongoing current account deficits. In contrast, the EGW data were produced by the simple procedure of scaling long-term debt by a constant factor derived from the ratio of total debt to long-term debt at some date after September 1989. It cannot be overstated how unreliable this latter approach is for the data series under review. It assumes, for example, no portfolio adjustments between short- and long-term debt after radical deregulation of the financial markets in 1984 and 1985. Further, it ignores that prior to October 1984 private overseas borrowings were restricted by law to long-term debt. Finally, it does not explain how the Reserve Bank survey the previous year failed to record the billions of dollars of short-term debt that the EGW series indicates had been built up after 1981. Thus the data series supplied to EGW must be discounted. Instead, the authoritative NZIER series indicates that total overseas debt was almost 30 percentage points of GDP higher in March 1995 than in March 1984 (assuming negligible domestically issued securities held by overseas residents in 1984). IV. New Zealand's Real Per Capita GDP At the heart of the decision to implement economic reform in 1984 was widespread dissatisfaction with the per capita growth rates achieved over the previous two decades. EGW present the following data to suggest that the reforms have had a positive impact on this statistic (p. 1872): Per capita growth (i.e., the increase in the ratio of GDP to the population of working age) proceeded at an average rate of 0.37 percent p.a. over the 1967 to 1984 period. The situation worsened over the first seven years of the reform period with per capita growth of just 0.04 percent p.a. Per capita GDP growth of 2.61 percent p.a. between June 1991 and June 1995 changed the situation, producing a 0.97 percent p.a. compound per capita GDP growth rate over the eleven years to June 1995. This was almost treble the per capita growth rate of the previous 17 years leaving the level of per capita GDP 6.8 percent higher in June 1995 that it would have been had the pre-1984 growth rate continued. There are three problems with these data. First, 1966/67 was a peak year in New Zealand's business cycle, coming at the end of several years of sustained growth before a collapse in the price of wool (a major New Zealand export commodity) triggered a sharp recession, whereas 1983/84 was a year of recovery after the previous year's downturn (see Figure 1 below). Since trend rates should be calculated either on the basis of cyclical averages or from cyclical peak to peak (Solow, 1997, p. 230), EGW's method introduces a downward bias to the pre-reform growth rate that should be corrected. Second, like all researchers of long-term trends in New Zealand, EGW were confronted with the need to construct the early values of their data series, because official quarterly data for real GDP began only in June 1977. Their approach was to use a series of real expenditure from March 1961 to March 1979 created by a Reserve Bank of New Zealand project reported in Grindell (1981). Conceptually, an expenditure series is not consistent with the intention to compare trends in output per working age person, and a production based index such as Statistic New Zealand's (1995, p. 127) historical series of real gross domestic product would be preferable. Third, in mid-1996 Statistics New Zealand released a major revision to their previously published real GDP statistics. The revisions resulted in significantly lower estimates of real GDP in recent years. To examine whether EGW's results are robust to correcting these problems, Figure 1 depicts real GDP per working age person for years ending March, using Statistics New Zealand's historical series for real GDP (INFOS series SNBA.S2AZAT) and a working age population series constructed from the INFOS series DPEA.SBEC, scaled by official estimates of the population aged 15-64 (Statistics New Zealand, 4 1996, updated). Early values of both data sets are available for March years only, and no attempt has been made to convert this series into year ending June data. Also shown in Figure 1 is a revised trend line, based on the annual growth rate between the two peak years of 1967 and 1982. (See end of file.) Figure 1. Real Per Capita GDP, New Zealand, 1960/61-1996/97 Figure 1 indicates that New Zealand's real GDP per working age person in 1967 was just under $29,000 (measured at 1991/92 prices), and was $32,400 in 1982, representing a pre-reform trend growth rate of 0.795 percent per annum. This is not high by international standards (although twice as high as the value reported by EGW), but Figure 1 shows that it is not true that the 1995 level of real per capita GDP lay above the value that would be predicted using the pre-reform trend. Instead, real per capita GDP was still 1.3 percent below trend in 1995. There is a further important observation to make from Figure 1. Consider the deep and extensive departure from trend that occurred between 1988 and 1994, before the economy returned close to trend in 1995. The economic sacrifice incurred during the transition phase of the reforms can be indicated by measuring this difference between the trend line and the actual output path. This amounts to $11,500 per working age person (in 1991/92 prices), or 32 percent of annual GDP in 1997. V. Conclusion New Zealand's program of economic reforms continues to attract international attention, and EGW's article is likely to be widely cited. Using official and public data sets, however, the previous three sections allow the following statements to be made in contrast to the claims of EGW. 1. Using an absolute poverty standard, the number of households below the poverty line increased from 4.3 percent in 1984 to 10.8 percent in 1993. 1. Between March 1984 and March 1995, New Zealand's total overseas debt rose by almost 30 percentage points of GDP. 1. New Zealand's real per capita output (based on working age population) in the year ending March 1995 was still 1.3 percent below the level that would have resulted had the economy continued to grow at the peak-to-peak trend rate of 0.795 percent per annum between 1966/67 and 1981/82. 1. The income sacrificed between 1987/88 and 1993/94 as a result of real per capita output being below its pre-reform trend was $11,500 per working age person (in 1991/92 prices), or 32 percent of annual GDP. These figures indicate why many economists in New Zealand remain cautious in our evaluation of the reforms. The program's close adherence to economic theory should have raised the economy's long-term potential. Nevertheless, the output sacrificed and the poverty created during the transition phase of the program were substantial, and it is still not possible to demonstrate a clear-cut improvement in measured growth performance thirteen years after the reforms began. 5 References Colgate, P. and J. Stroombergen (1993). A Promise to Pay: New Zealand's Overseas Debt and Country Risk. Wellington: New Zealand Institute of Economic Research, Research Monograph 58. Easton, B. (1995). 'Poverty in New Zealand: 1981-1993.' New Zealand Sociology, 10(2) : 182-213.. (1996). 'Income Distribution.' Chapter 4 in B. Silverstone, A. Bollard and R. Lattimore, Eds. A Study of Economic Reform: The Case of New Zealand. Amsterdam: North-Holland. Evans, L., A. Grimes and B. Wilkinson with D. Teece (1996). 'Economic Reform in New Zealand 1984- 95: The Pursuit of Efficiency.' Journal of Economic Literature, 34(4) : 1856-902. Grindell, D. (1981). 'Consolidated National Accounts for New Zealand on an SNA Basis.' Research Paper No. 32. Wellington: Reserve Bank of New Zealand. Kelsey, J. (1997). The New Zealand Experiment: A World Model for Structural Adjustment? Second Edition. Auckland: Auckland University Press and Bridget Williams. Krishnan, V. (1995). 'Modest but Adequate: An Appraisal of Changing Household Income Circumstances in New Zealand.' Social Policy Journal of New Zealand, 4 : 76-97. Silverstone, B., A. Bollard and R. Lattimore (1996), Eds. A Study of Economic Reform: The Case of New Zealand. Amsterdam: North Holland. Solow, R.M. (1997). 'Is There A Core of Usable Macroeconomics We Should All Believe In?' American Economic Review, 87(2) : 230-32. Statistics New Zealand (1995). New Zealand System of National Accounts 1995. Wellington: Statistics New Zealand.. (1996). Demographic Trends 1996. Wellington: Statistics New Zealand. Stephens, R., P. Frater and C. Waldegrave (1995). 'Measuring Poverty in New Zealand.' Social Policy Journal of New Zealand, 5 : 88-112. 6 7