Monday, May 5, 2014

Savings, interest rate and the exchange rate in New Zealand

The New Zealand’s opposition Labour Party announced that in order to lower average real interest rates and relief pressures on the unusually appreciated currency, they would pursue a policy of to increase mandatory savings. The details of the policy have not been worked out yet, but the gist of the argument is that higher savings lead to lower real interest rate on average, which would lower the real exchange rate (real depreciation of the currency). In general, there is no prima facia problem with advocating a mandatory savings policy. Presumably, if people perceive the policy as permanent, the stock of domestic capital would increase, consumption would fall, and diminishing marginal returns to capital would reduce the equilibrium real interest rate in the long run.[1] But there are three issues here:

First, New Zealand is an open economy, regardless of what measure we use for openness. When Kiwis want to consume more today than tomorrow they would borrow from abroad so it does not necessarily follow that consumption will fall because of a mandatory savings policy. New Zealand is a stable Western democracy and our robust institutional arrangements such as our laws guarantee that Kiwis would not default on their international obligations.

Second, we have been concerned about the relatively high real interest rate in New Zealand for a long time, but we do not have a firm knowledge why is that. New Zealand’s inflation-indexed interest rate is more than one percentage point higher than that in the U.S. The data from May 1998 to-date show that the average difference between New Zealand’s rate and the U.S. rate is 1.08 percent. To be precise, the inflation-indexed interest rate for New Zealand has been above the U.S. rate at every moment in time.[2] What explains such a persistent difference? A recent paper in the Journal of Finance provides a new theory and test.[3] In short, it says that the relative sizes of the economies matter. The U.S. economy is massive relative to New Zealand, therefore, U.S. risk-free bonds are relatively more expensive than New Zealand’s, and hence the average U.S. real interest rate is relatively lower. The author tests a large cross sectional data and shows that differences in the sizes of economies explain a large fraction of cross-section variation in currency returns. The proposed Labour Party policy would be ineffective in this case. This means that, on average, we will not be able to have a lower real interest rate than the United States, and most other sizable trading partners.

Third, in theory, relative productivity growth differentials or the covered and uncovered interest rate parities explain the exchange rate depreciation rate. But reality is different from theory when it comes to explaining the exchange rate. Those commonly used exchange rate models do not work well in practice. Parity conditions are among the six known puzzles in international economics.[4] The magnitude of the expected change in the exchange rate, nominal or real, which we observe in New Zealand cannot explain the interest rate differentials between New Zealand and the U.S. so we cannot be sure that a policy, which aims at lowering New Zealand’s equilibrium real interest rate, guarantees a depreciated currency.

In summary, the theoretical linkages between savings, real interest rate and the real exchange rate one hand, and the empirical evidence on the other, are not as clear as the Labour Party leader think they are.

Finally, let me say a few things about the real exchange rate, which might help understand the problem. One way of looking at the real exchange rate is to examine the ratio of the price of non-tradable goods to the price of the tradable goods. This ratio has been increasing because non-tradable prices have been relatively higher than tradable prices since 2006 (see the RBNZ data files online). Most of the prices of tradable goods and services are beyond our control because New Zealand is a price taker. So the real exchange rate appreciation issue, which policy might be able to influence, is related to the price of non-tradable goods. The largest components of non-tradable include health, education, housing, energy & water, and telecommunications. The demand for these goods and services keeps increasing because of the increase in population and income, etc. So prices are expected to keep increasing. Examination of the available input-output tables of the New Zealand economy (1996 and 2007) suggests that mining; energy & water and construction have experienced increasing cost per unit of output, growing output, and declining productivity – the Baumol disease.[5] Mining aside, because it is a tradable good sector, therefore, some of the increases in non-tradable prices, which cause the real exchange rate appreciation, might be related to some imbalances in the energy & water and the construction sectors. Although the shares (weights) of value added / total New Zealand's output of these two sectors are relatively smaller than those of agriculture and manufacturing, the share of construction’s value added in total output might have doubled between 1996 and 2007 input-output tables. The Policymaker may want to ask: how much of the rise of the non-tradable price is associated with the rising marginal costs in these two secorts; why productivity in these sectors have not increased; and what could be done to increase it because if the price of non-tradable goods and services continue to increase by more than the prices of tradable goods and services, the real exchange rate will continue to increase. 

[1] Wicksell’s equilibrium or natural real interest rate is the rate of returns on the economy’s capital stock.

[2] The period following the global financial crisis to-date exhibited a significant reduction of the U.S. interest rate.

[3] Tarek A. Hassan, (2013), Country Size, Currency Union, and International Asset Returns, Journal of Finance, Vol. 68, Issue 6, 2268-2308.

[4] Obstfeld, M. And K. Rogoff, (2001), ‘The Six Major Puzzles in International Macroeconomics: Is There a Common Cause?’ in NBER Macroeconomics Annual 2000, Volume 15, eds., Ben S Bernanke and Kenneth Rogoff, MIT Press.

[5] Baumol, W. J., (1968), Productivity Growth, Convergence, and Welfare: What the Long Run Data Show, American Economic Review, Vol. 76, No. 5, 1072-1085.