Cycles and Depressions in New Zealand History
This is a revised version of a paper given to IPS seminar, 22 November 2008. It is similar to the article published in Policy Quarterly, Volume 3, No 1, February 2009, pp.17-23.
Keywords: Macroeconomics & Money; Political Economy & History;
Introduction
It is said that a banker’s most dangerous expression is ‘It’s different this time’, a phrase used to justify repeating actions which proved mistaken in the past. The economist’s approach is a little different. Ideally, the same basic model should apply on all occasions. However each time its application differs because
– the parameters are different;
– the external shocks are different;
– the policy institutions (and therefore their reaction functions) are different.
As we face what may be the greatest financial crisis which New Zealand has ever experienced – albeit not necessarily the greatest economic crisis – we can try to use the experiences of the past to identify the underlying issues and the potential policy responses.
New Zealand has experienced cyclical economic downswings since it became a market economy two hundred years ago. Not all are particularly well evaluated. Perforce, this paper confines itself to those for which there is sufficient scholarly work. I start with short-run cyclical fluctuations, then turn to the three major periods of long depression in New Zealand’s post-1840 history.
The Standard Post-War Business Cycle
The best studied short-run fluctuations are those of the post-war business cycle. Recent work by Viv Hall and John McDermott (2007 and forthcoming) identifies 11 cycles (excluding the current one), which implies an average length of about five years. Others including myself (Easton 1980, 1997) have identified up to perhaps another six smaller cycles, which would reduce the average to three years. So there is some disagreement about what constitutes a business fluctuation, and they are not very regular.
That means they are not simply the theoretical ‘trade cycles’ described in the seminal paper by Paul Samuelson (1939), which showed that under certain circumstances interaction between the orthodox consumption multiplier and the investment accelerator generates a regular sine wave.
The irregularity of New Zealand’s business cycle tells us that while Samuelson’s mechanisms may be important, the New Zealand business cycle is not based on a self-sustaining process but is likely to be a consequence of external shocks, as in the model identified by Ragnar Frisch (1933). The impression is that the mechanism is strongly damped: each shock generates a single fluctuation, with subsequent ones obscured by the next shock. The irregularity also tells us that the New Zealand business cycle cannot have been simply induced by the an election cycle.
Nor are the fluctuations a regular sinusoid about a trend. Rather, the peak phase tends to be a growth plateau of four to twelve quarters, so it is difficult to identify a peak. The trough is much sharper – it can usually be identified to within a three month period. This suggests that the characteristic shock induces a downturn, from which the economy recovers to running at its growth capacity until another shock ensues. This has some of the characteristics of the theory of the trade cycle proposed by John Hicks (1950).
Crucially, the business cycle and growth trend are not independent of one another, but are intimately linked. The longer the economy is on the growth plateau, the stronger the trend growth rate.
The best account of the cycle mechanism in New Zealand is still that of Eric Haywood and Colin Campbell (1976, 1978), written in the pre-1985 fixed-exchange-rate environment. The external shock which precipitated the typical business downswing was a fall-off in the price of exports (relative to import prices). The export sector suffered a loss of revenue, which caused both a reduction in spending (including on investment) and rising borrowing from the banks and other sources of finance. But the lending capacity of the banks was limited because they experienced a drain of foreign funds, as receipts from exports were lower than the outgoings for imports and debt servicing.
Because the exchange rate was fixed this deficit of foreign funds had to be covered by government offshore borrowing. The government, facing falling receipts itself, would reduce its spending, adding to the downward pressures from the export sector. Domestic sectors would face reductions in spending and so there would be a general contraction in production. At some stage these processes would reach a lower limit – as they do in the Samuelson and Hicks accounts of the business cycle – and the domestic upswing would begin, especially if there was a recovery in the terms of trade.
Here I make three modifications to the Haywood-Campbell analysis. First, there is an inventory cycle which is shorter, but of considerable amplitude, which exaggerates the underlying shock. Second, Schumpeterians would argue that the downswing was important in extinguishing less productive firms which survived during the boom. The released resources – labour and capital – would be transferred to more productive firms in the upswing.
The third caveat is more complex. Haywood and Campbell were writing before the floating exchange rate. Under the fixed-rate regime, banks facing a shortage of foreign funds obtained them from the Reserve Bank; hence the government’s need to borrow foreign currencies. Today we have a floating exchange rate, and in principle the Reserve Bank need not borrow offshore.
How the economy fluctuates under a floating exchange rate has not been properly evaluated; the long rogernomics recession involved some other events which complicate the analysis, and the fact that the Australian economy sailed through the Asian crisis while the New Zealand economy did not suggests that policymakers got it wrong. (The downswing, troughing in first quarter 1998, involved an unfortunate monetary policy response guided by a Monetary Conditions Index which was mathematically wrong, without theoretical foundation and empirically untested.)
In summary, the postwar business cycles all display the same internal economic mechanisms, although the external shocks may have differed, and different policy mistakes may have been made. But monetary phenomena were endogenous – albeit in different ways reflecting different institutional arrangements – both before and after the floating of the exchange rate. This endogeneity of money was not true for the first two major downswings New Zealand experienced before the war.’
The Long Depression of the 1880s
There were short-run fluctuations throughout the nineteenth and early twentieth century. Some were externally generated, as in the late 1840s which left Maori stranded with provisions for ships which did not arrive because of a financial crisis in Britain. Some were home-generated downturns which occurred when temporary unsustainable activities ran out of resources – the whale or alluvial gold quarries – or a boom from war or borrowing came to an end, or when settlements ran out of their initial capital for investment.
However the Long Depression of the 1880s and the Great Depression of the 1930s involved more complex processes under monetary arrangements which were different from the post-WWII ones. There was then no central bank in New Zealand. Rather the trading banks held sterling reserves. A run-down of these balances (as when export revenue fell relative to import revenue) led the banks to contract domestic lending. New Zealand was on a ‘sterling’ standard, not a gold standard.
In October 1878 the City Bank of Glasgow failed with a capital deficiency for its liabilities of £12.4m were covered by only £7.2m of assets. There had been deliberate falsification in the accounts, and all the directors were jailed. There being unlimited liability, about 1000 shareholders were ruined. (What must have been exceptionally galling was that so skillful were the deceptions that on the last day £100 shares were trading at £236.)
New Zealand was affected in two ways. First, the Bank had made speculative investments in Australia and New Zealand, in the hope of recouping banking losses. Second, and following three further bank collapses in December, there was a tightening in the London money market. New Zealand had spent the previous decade relying on borrowing in London to support the Vogel boom. The tap was turned off and there was a credit contraction. Trading bank advances, which had almost trebled between 1870 and 1879, fell 15 percent in the following year, and while there was some subsequent growth, New Zealand struggled through the next decade in ‘The Long Depression’.
There are two other elements crucial to this story. First wool prices fell. Although they had been falling since 1873, the further fall of a third between 1878 and 1886 reduced the banks’ sterling receipts and the income of farmers (and of those whom they purchased from). So while the Long Depression was precipitated by a monetary crisis overseas, the independent terms of trade deterioration compounded the misery.
Second, there had been land speculation in the 1870s, including irresponsible lending by banks (as the City of Glasgow Bank story suggests) but also fueled by the government borrowing of the Vogel era. As a result land prices were out of line with the returns from farming them. As a bank inspector (W.G. Rhind) commented about Hawkes Bay, the land prices were ‘enormous’ and fictitious’, and ‘most monstrous payments’ were being made, thanks to settlers holding ‘the most sanguine, not to say wild ideas of the importance of the place’.
Owners were thus saddled with excessive interest payments on overvalued land. Banks were faced with the dilemma of carrying such owners, or bankrupting them and taking over the land, possibly unable to onsell it, and probably having difficulty maintaining its productive value without a committed farmer. Some banks later failed. But because they were a peripheral part of the London banking system there was no great financial shock, although the New Zealand Government bailed them out.
Factor and product prices were flexible in those days, and there was a general lowering of price levels. However debts are usually set in fixed nominal terms, and so are inflexible. One of of the greatest problems in each depression has been how to realign debts with actual prices; sometimes bankruptcy is the only option.
The Great Depression of the 1930s
The toxic combination of international monetary turmoil, slumping export prices, and excessive domestic debt levels was repeated in the Great Depression of the 1930s. This time the monetary crisis almost certainly contributed to depressing the terms of trade.
My analysis suggests that the Great Depression was in fact part of a longer Interwar Depression, but I won’t pursue that here. (Easton 1997) Certainly the debt problem was prior, for farm land prices were too high. In the 1922-9 period the return on farmer equity averaged 3.4 percent p.a. whereas they were paying 6.5 percent p.a. on farm mortgages; retirement would have given them better return. Future productivity and prices lifting incomes might have changed the balance. Such optimism proved illusory. In the four years after March 1930, the returns on farming did not even cover the value of the labour of the typical farmer, let alone give a return on the capital.
The fall in farm output prices (even relative to input prices) was caused by the world economic crisis which had been precipitated by the 1929 Wall Street Crash. (It is now generally accepted that the causes of the crisis were more complex than the crash by itself.) It is harder to stop farm production than factory production – harder to layoff livestock than the workforce – so there tends to be an oversupply of farm products relative to manufactures, which drives down their relative price. With land prices – or, more accurately, the farm debt secured on land – already too high, the fall in output prices was a recipe for disaster.
The government position was not wholly robust either. During the 1928 election campaign, Joseph Ward promised to borrow £70m. He probably intended to do so over ten years, and so was nominating a rate not unlike that of the previous decade. The borrowed funds were used for the works program. When the money dried up, the programs were cut and workers laid off.
Government borrowings in the London financial market were first deposited with the Bank of New Zealand (the trading bank the government partly owned), and became part of its sterling balances. The money used in New Zealand which ended up in one of the other five trading banks would be transferred to their London sterling balances. Funds used to purchase imports would end up in a foreign trading bank, reducing the aggregate sterling balances in London. Thus the Government’s sterling loan would be run down, initially to other New Zealand banks, later to those outside the New Zealand system.
Similarly export receipts would end up in one of the local bank’s London accounts, be transferred to other banks’ accounts as the exporter spent the money in New Zealand, and drain away as the money was outlaid on imports.
So long as banks had a surplus of sterling balances they could make advances to New Zealanders, knowing that the surplus would be drained to other New Zealand banks and, eventually, outside them. If there were no additional funds coming in from government loans, and if export receipts fell (as they did by over a third between 1929 and 1931 from £55m to £35m in the year), the banks would have to contract their advances.
On top of the reduction of expenditure because farmers had less income, there was a reduction in funds for investment (including on public works) while trade credit was harder to obtain. The same economic mechanisms which we later see in the post-war business cycle generated an economic downswing. The fall from the peak in the year ended March 1930 to the trough in 1934 amounted to around 12% in real GDP terms (and more than double that relative to the trend rate of growth).
It is common to criticise the policies of the government of the early 1930s. Perhaps they were not fair, failing to share the burden of adjustment equally. The social hardship they caused is well recorded; the worst hardship was from unemployment. Some of the criticisms are extravagant. It is true that wages were cut by government fiat, but they followed prices down and real wages were broadly maintained.
The government did not pursue expansionary measures, for without sufficient control over the monetary system they could not practice macroeconomic management. Indeed they had so little influence that the foreign reserves (sterling balances) of the trading banks rose between 1930 and 1935. Unlocking the reserves for public policy purposes required a central bank.
New Zealand had entered the Depression with excessive debt, and the fall in both export and import prices disrupted the relationship between external and internal prices. Much of the policy activity of the period was to regain the balances between debt and domestic prices and within the price system. The devaluation of the exchange rate, the cutting of wages, the Mortgagors and Tenants Relief Acts all had this purpose. This realignment, plus new policy institutions such as the Reserve Bank, provided the foundation of the expansion for which the subsequent Labour Government is renowned.
This realignment of nominal price relativities is central to a sustainable Keynesian expansion. The conventional account of New Zealand in the Great Depression is based on a simple model in which a single commodity can be expanded and contracted by demand management. However an open economy must have multiple commodities, for otherwise it would not be necessary to export and import. A multi-sectoral analysis in which the relative prices between sectors and the debt they carry is a critical part of the story of the Great Depression. Fortunately the economist advisers of the 1930s grappled with it.
The Third Depression 1966-93
Describing the period from 1966 to 1993 as a ‘depression’ may seem artificial, It certainly was a period of high unemployment, and poor economic performance. What distinguishes it from the earlier depressions is that the world economy prospered so New Zealand grew slower than they generally did. It may be better to describe it as a period of slower economic growth – even a climacteric or retardation in the trend of economic growth – or a structural adjustment responses to a new world situation (poorer prices for New Zealand’s key exports). It is included here because some commentators have the compared the period – especially the Long (Rogernomics) Recession of 1986 to 1993 – to the current economy situation. This depression began with the fall in the auction price of wool in December 1966. It was a structural change – except for the 1971-2 world commodity boom, the relative wool price never recovered. The price shock came not from some international monetary crisis but was particular to the industry: competition from synthetics.
As about two fifths of export receipts in the 1960s came from wool, a structural price fall of around 40 percent meant a 16 percent loss of export receipts. It was actually worse than that, because in 1966 another 20 percent of receipts came from sheep-meats, so the sector which dominated exports was devastated. (Today sheep meats contribute about 15 percent of the receipts from the exports of goods and services, with wool providing a further 5 percent.)
The permanent price collapse meant that much of the capital value in the industry had to be written off. The downswing proved longer than usual, which inhibited physical investment and innovation. As the rest of the world economy prospered, New Zealand’s stagnated, and so it declined in the world rankings of income per head. Over the decade after 1966 the New Zealand GDI per capita fell about 15 percent relative to the rest of the OECD. Following the downswing of the later 1970s, there was a temporary recovery from the world commodity boom in 1971-2 which switched over into the sharp downswing of the 1973-4 oil shock; New Zealand followed. The economy was weak in the late 1970s and unemployment more apparent.
While the standard economic mechanisms of all the early postwar downswings occurred, the structural price collapse in the economy’s leading sector triggered a major long-term restructuring of New Zealand’s external economy. By the end of the 1970s it had moved from being one of the most concentrated exporters, by both product and destination, to the middle of the OECD pack. However the domestic restructuring was slower. The available economic analysis was only of limited help, because it was trapped in a single commodity model. (That epitome of conventional wisdom, the Planning Council, appears to have never grasped the significance of the terms of trade.) Instead the question of who was to bear the burden of the deficit was resolved by ongoing inflation, which allowed each group in the economy to shift the burden temporarily elsewhere. Ultimately it was the fixed interest investors who suffered.
The dominant politician of the day, Robert Muldoon, his childhood scarred by the Great Depression and naturally conservative, found it politically easier to protect weak industries than to have their workers redeployed into more productive enterprises.
So for over two decades there were two largely unanswered questions:
– How was the reduction in effective incomes to be shared among the different parts of the economy?
– How were the domestic labour and other resources to be redeployed (noting that the diversification muddied the distinction between the external economy and domestic economy which had previously existed)?
Growth was stronger in the early 1980s (partly as a result of the ‘Think Big’ major projects, which proved less valuable than expected, when oil prices fell in the mid 1980s). Then the ‘Rogernomics’ policy regime of almost unconstrained market liberalisation heralded a ‘long recession’ from 1987 to 1993.
The new regime addressed the outstanding questions. In the process of quenching inflation it reduced real benefits, and wages at the lower end of the pay spectrum. Workers locked up in the protected firms were dumped onto the labour market. In the 3¾ years from 1988, the equivalent of 43 percent of the labour force registered with the Department of Labour as unemployed (multiple registrations not included).
While the changes involved some fundamental modifications to employment and government spending arrangements, monetary arrangements were central. Following the floating of the New Zealand dollar in March 1985, the trading banks purchased their foreign currency directly on the market rather than from the Reserve Bank as had been the rule in the fixed exchange rate regime. The exchange rate varied depending on the supply and demand for currencies and the expectations of their future. Export profitability is the inverse of the (real) exchange rate, so a high rate means low export returns, especially following the withdrawal of the subsidies. As the Government allowed the real exchange rate to rise following the 1984 devaluation, export growth slowed down. The New Zealand economy downsized with per capita GDP falling in each of six years, costing another 15 percent of per capita GNI relative to the OECD.
In the floating regime borrowing was an individual matter and its wider implications were ignored. There was, for example, no certainty that the exchange rate would support the export sector sufficiently to ensure that the economy could grow. The consequences for the export sector were evident as early as 1985. A high exchange rate was key to the anti-inflation strategy, since the cheaper imports pressed down on import competing sectors – most of which closed down – while raising real wages without raising nominal wages, for those who still had jobs.
Foreign borrowing fueled a speculative financial bubble which burst in October 1987 when the world’s share markets collapsed. This was a financial rather than a monetary crisis, and New Zealand’s monetary system was not seriously compromised. However the local share market failed to recover with any of the vigor of its counterparts, demonstrating that it was artificially stimulated by the market liberalisation and the easy borrowing. There was no underlying strength or robustness in an economy whose tradeable sector was being undermined by the same high exchange rate which was a result of the capital inflows which fueled the speculation.
In summary, then, New Zealand’s third depression was a response to price misalignment, initially from the collapse in the wool price, but subsequently from an over-valued exchange rate. Debt was a problem insofar as it reflected outdated price relativities, but except in the 1986-7 financial boom and bust it was not a major factor. It is hard to blame the rest of the world for New Zealand’s difficulties in this case; local policy responses certainly exacerbated them.
Is it the Same This Time?
Table 1 compares the three substantial depressions which New Zealand has faced.
Table 1
This paper’s aim has been to identify the key elements which contributed to previous New Zealand depressions. It turns out that while many of the economic mechanisms were the same, there were differences in external circumstances, terms of trade, debt and economic management. Closer inspection suggests the first two depressions were similar, but the third was different. It is included here because its institutional arrangements are closer to today’s circumstances and because it is more prominent in the popular memory.
How do those previous experiences compare with what we are now heading into?
External Circumstances
There is no doubt that the world is facing a monetary crisis, and it is probably optimistic to assume it will be soon over, for there are almost certainly consequential collaterals which have yet to work their way through. Arguably the world is better placed to deal with the crisis than on previous occasions; arguably this crisis is more widespread and complex than on previous occasions.
The world economy has entered a recession. The late 2008 central forecasts suggested the world recovery will begin by early 2010. That may be optimistic. Nor can we rule out that the subsequent expansion will be weak in output terms or highly inflationary, or perhaps different in ways we cannot anticipate.
Price Alignments
There has been some falling off in commodity terms of trade from recent high levels. Some of the falls, such as for oil, are of benefit to New Zealand. There is no reason to believe that the current export price trends are a detrimental structural change like they were in 1966 (except that we might expect energy prices to rise). There is reason to believe that the food price terms of trade may be secularly rising.
Assessment of the domestic price structure is complicated by the real exchange rate, which undergoes medium term cycles under the current macroeconomic policy framework. The rise in the 2000s choked off export growth and tipped the New Zealand economy into a growth slowdown – even a recession – before the world recession started. There has been a subsequent fall in the exchange rate in part as the cycle enters the next phase, but probably also because of capital flow responses as a result of the world monetary crisis. The exchange rate is near the middle of its ‘normal’ range whereas recently it was at the top. It seems likely to fall further as the cycle continues.
We do not know how quickly the export sector will respond to this more favourable exchange rate. It almost certainly faces a reduction in world demand which will impact on different sectors quite differently.
The fall in the exchange rate suggests there may be upward movements of domestic prices, although these may be moderated by the fall in world commodity prices.
Debt
By the standards of previous depressions the government and the business sectors appear to have favourable debt levels. However the household sector is holding unusually high debt by past standards. This is largely secured against over-priced housing, although there is some consumer debt with little security. It is almost all owed to banks. Insofar as they will be protected by the Reserve Bank, the private sector debt becomes a public sector problem.
While most consumer debt is legally secured against housing, it is largely serviced from labour earnings. As long as unemployment remains low, the housing debt problem is manageable for most individuals, although there are some who are over-borrowed against their human capital. The challenge will be at the macro-level; can New Zealand roll over its international debt at reasonable cost?
Mention has to be made of the non-bank financial sector. It would appear that much of it was over-borrowed, invested in schemes which would never give an adequate return except on implausible assumptions of capital gains. Much of that has been wiped out with defaults but the offset is many households have lost substantial deposits, and are the poorer.
Government Macro-economic Management
Its current quality is too soon to tell.
Conclusion
Table 2 puts the previous sections discussions in the context of Table 1.
Table 2
In conclusion: Yes this time it is the same; yes, this time it is different.
References
Campbell, C. and Haywood, E. (1978) The International origins of New Zealand’s Business Cycle, Welington: NZIER Research Paper No 36.
Easton, B.H. (1980) ‘Three New Zealand Depressions’, in W. E. Willmott (ed) New Zealand and the World: Essays in Honour of Wolfgang Rosenberg, Christchurch: University of Canterbury.
Easton B. H. (1997) In Stormy Seas, Dunedin: Otago University Press.
Frisch, R. (1933) ‘Propagation Problems and Impulse Problems in Dynamic Economics’, in Economic in Essays in Honour of Gustav Cassel, London: Allen & Unwin, London.
Hall, V.B. and McDermott, C.J (2007), ‘A Quarterly Post-World War II Real GDP Series for New Zealand’, Motu Working Paper 07-13, http://www.motu.org.nz/publications/workingpaperseries/.
Hall, V.B. and McDermott, C.J. (2006), The New Zealand Business Cycle, Working Paper 21/2006, Canberra: Center for Applied Macroeconomic Analysis (CAMA), Australian National University, http://cama.anu.edu.au/publications.asp; revised version forthcoming in Econometric Theory.
Haywood, E., & C. Campbell (1976) The New Zealand Economy: Measurement of Economic Fluctuations and Indicators of Economic Activity, 1947-74, RBNZ Research Paper 19.
Hicks, J.R. (1950), A Contribution to the Theory of the Trade Cycle, Oxford: Oxford University Press.
Samuelson, P.A. (1939) ‘Interactions between the multiplier analysis and the principle of acceleration’, Review of Economics and Statistics, 21.
Sinclair, K. & W. Mandle (1961) Open Account: a History of the Bank of New South Wales in New Zealand, 1861-1961, Wellington: Whitcombe and Tombs.