RDP 2014-11: Exchange Rate Movements and the Australian Economy 4. Scenario Analysis

Having established the aggregate and industry implications of exchange rate shocks, we now use these results to ask how exchange rate movements, which are combinations of shocks and endogenous responses to economic developments, have contributed to Australia's economic performance over the past decade. To do this, we construct two scenarios. In the first, we discuss how the Australian economy might have behaved if it had experienced the same international and domestic shocks but the nominal exchange rate had not appreciated. In the second, we examine how the Australian economy might have evolved without the terms of trade boom and the global financial crisis.

4.1 What if the Nominal Exchange Rate Had Remained Constant?

Starting in the March quarter of 2003, we examine a scenario whereby all of the model's structural shocks, except the one for the exchange rate, remain at their estimated historical values and impose shocks on the real exchange rate in such a way that the nominal exchange rate remains unchanged at its initial level. To do this, we make use of the definition of the real exchange rate, which implies that:

Where ΔNTWI_t is the percentage change in the nominal exchange rate, ΔRTWI_t is the percentage change in the real exchange rate, π_t is the domestic inflation rate and is the overseas inflation rate. By assumption, the foreign inflation rate is unaffected by domestic economic conditions and the domestic inflation rate is only affected by real exchange rate shocks with a lag. Consequently, by imposing a suitable real exchange rate shock each period, given overseas inflation (which is unaffected by Australian economic developments) and domestic inflation (which we allow to evolve endogenously), we are able to simulate a path of the real exchange rate for which the nominal exchange rate remains constant.

Figure 6 shows the counterfactual paths of the real exchange rate, inflation, the cash rate and GDP growth in this scenario.^[15]

The model suggests that, even if the nominal exchange rate had remained constant, the economy would have experienced a large real exchange rate appreciation over the past decade. This reflects the fact that most of the appreciation of the real exchange rate was due to economic fundamentals, in particular the large increase in the terms of trade. A flexible nominal exchange rate ensured that this real exchange rate appreciation occurred in a timely manner, and mitigated much of the effect of the commodities boom on domestic prices. But even without this mechanism, the model suggests that, in a world of rising commodity prices, the real exchange rate would still have appreciated eventually.

However, the model also suggests that, without the appreciation of the nominal exchange rate, the terms of trade boom would have had a destabilising effect on the nominal side of the economy. In the scenario, the inflation rate increases to over 6 per cent in 2008 and remains at an elevated level subsequently. This reflects the fact that, if the nominal exchange rate is fixed, a real appreciation requires an increase in the domestic price level relative to the foreign price level. If foreign inflation is unaffected by Australian economic conditions, this implies higher domestic inflation. The model suggests that Australian nominal interest rates would have been considerably higher if the nominal exchange rate had not appreciated, reflecting higher domestic inflation.^[16]

Initially, the model suggests a constant nominal exchange rate would have increased the pace of economic growth. This is because, in the scenario, the appreciation of the real exchange rate occurs more slowly than it did in reality, as domestic prices adjust more slowly to economic shocks than nominal exchange rates do. Once the real exchange rate begins to appreciate, however, GDP growth in the scenario is similar to its actual level and, if anything, is slightly lower during the global financial crisis, as higher inflation limits the ability of the RBA to respond to weaker domestic economic conditions.

An important caveat to this result is that it reflects the impact of a sequence of unanticipated negative shocks to the nominal exchange rate and hence may overstate the impact of a persistently weaker nominal exchange rate on inflation. If the RBA had anticipated that the terms of trade boom would not have been accompanied by an appreciation of the nominal exchange rate, it may have responded more aggressively to rising inflation than in this scenario. This would have moderated the increase in inflation but also reduced the pace of economic growth.

Figure 7 shows the estimated implications of a constant nominal exchange rate for various sectors of the economy. As suggested by the aggregate responses, several industries would have experienced more rapid growth in the early stages of the mining boom if the nominal – and, by association, real – exchange rate had not appreciated. For the business services and personal services industries, the model suggests that this faster growth in the initial stages of the boom would have left the level of output in these industries noticeably higher today. However, for other industries, such as mining and construction, a constant nominal exchange rate would have had left the level of output little changed, while output in the goods distribution sector would have grown less rapidly.

The manufacturing sector would also have benefited from a weaker Australian dollar in the early stages of the terms of trade boom. However, even a constant nominal exchange rate would not have prevented the ongoing decline in the relative size of manufacturing, which is due to broader technological and economic factors. At most, according to this scenario, a weaker exchange rate would have reduced the pace of this decline.

4.2 What if There Were No Foreign Surprises?

As a second exercise, we examine the contribution of foreign developments to the appreciation of the real TWI and other domestic economic outcomes over the past decade. To do so, we run a second simulation in which we set all domestic shocks, including the exchange rate shocks, at their estimated historical values, and impose shocks on the foreign variables such that US output grows at its estimated trend and the terms of trade is constant. In this way, the scenario abstracts from the terms of trade boom and the global financial crisis as well as other, less notable, foreign developments.

Figure 8 suggests that, absent unusual foreign circumstances, the real exchange rate would have remained at around the same level in the 2000s as it did during the previous two decades. That is, the appreciation of the real exchange rate over the past decade has been almost entirely due to global factors. However, despite the weaker exchange rate, output would have grown more slowly and inflation would have been lower in the mid 2000s than was actually the case. Consequently, the cash rate would also have been lower, although the model suggests that this would not have been sufficient to offset fully the absence of the terms of trade boom.^[17]

Since the onset of the global financial crisis in mid 2008, the Australian economy has faced a more challenging international environment. Weak economic growth in advanced economies has exerted a contractionary effect on the Australian economy. However, growth in emerging economies, which has been associated with ongoing strength in commodity prices, has supported domestic economic activity. The model confirms that the slowdown in advanced economies lowered Australian economic growth in 2008 and 2009. However, it also suggests that, on balance, international economic developments have had a neutral impact on Australian economic activity since 2010.

In terms of the nominal variables, the model suggests that, absent international developments, inflation would have also been higher in recent years. But, unlike the previous scenario, inflation would have remained broadly in line with the RBA's inflation target. Higher inflation and stronger economic growth would have translated into higher nominal interest rates. However, at 5–6 per cent, the cash rate would have been at around its average level in the inflation-targeting era.

Figure 9 shows the sectoral implications of the no foreign surprises scenario. The manufacturing, goods distribution and personal services sectors would all have grown considerably more slowly in the mid 2000s without the income boost from the early stages of the terms of trade boom and the strong economic conditions in other developed economies. This result coincides with the findings of Downes, Hanslow and Tulip (2014) that, in the early stages of the terms of trade boom, the demand-side effects of rising mining investment largely offset the competitiveness effects of a stronger exchange rate for many import-competing domestic industries.

More recently, global economic conditions have had a negative impact on output in these sectors. In the case of manufacturing, this may reflect the impact of subdued demand from advanced economies, which account for a larger share of Australia's manufacturing exports than they do of resource exports. For the services sectors, the result may reflect the reductions in domestic income and wealth resulting from the global financial crisis as well as the effects of lower levels of business and consumer confidence that these events, and their aftermaths, generated.

The model suggests that global economic developments had a comparatively small effect on the quantity of mining output, especially in the early years of the scenario. At first glance, this might seem surprising as higher commodity prices would be expected to have triggered an increase in mining production. However, it is consistent with the idea that the early stages of the terms of trade boom led to income gains without large increases in production as firms waited to see how persistent the boom was before committing to large investment projects – and that these investment projects take several years to result in increased capacity (Plumb, Kent and Bishop 2013).^[18] Indeed, the model suggests that, more recently, mining production would have been lower without the cumulative effect of the global economic developments of the past decade.

Putting the constant nominal exchange rate and no foreign shocks scenarios together, the results suggest that foreign economic developments had a substantial positive impact on Australian economic activity in the mid 2000s. Absent these developments, the economy would have experienced a period of below-trend growth, partly mitigated by more accommodative monetary policy and a modest exchange rate depreciation.^[19] The appreciation of the nominal exchange rate limited the spillover of increased economic growth and higher domestic incomes resulting from the terms of trade boom onto the nominal side of the economy. Without this appreciation, our model suggests that much tighter monetary policy would have been required to maintain a rate of inflation consistent with the RBA's inflation target.

Since 2008, foreign economic developments have had, overall, a contractionary effect on the Australian economy – particularly during the global financial crisis. Absent these developments, nominal interest rates would have been at more normal levels, although the effect on economic growth in recent times would have been modest.

Although the no foreign surprises scenario featured a weaker real exchange rate than the constant nominal exchange rate scenario, output growth was no faster in the former scenario and inflation was lower. The differing outcomes reinforce a key message of this paper: that the impact of an exchange rate movement on the Australian economy ultimately depends on what factors cause the movement.

Footnotes

Note that our assumption that Australian variables do not affect US GDP or the terms of trade means that these foreign variables are unaffected by this scenario. [15]

At first glance, this increase in inflation might seem large. However, it is of broadly similar magnitude to that which occurred between 1972 and 1974, which was an episode in which Australia experienced a large increase in its terms of trade in an environment of limited exchange rate flexibility (Atkin et al 2014). [16]

Note that this exercise understates the effects of the terms of trade boom on domestic living standards because, as well as increasing GDP growth, the boom also increased domestic income. [17]

In addition, some of the increased demand due to the commodities boom may have been met by other industries, including the construction, business services and manufacturing industries (Rayner and Bishop 2013). To the extent that the contribution of these industries was particularly large during the investment phase of the mining boom it is not surprising to see a delayed response in mining output to higher commodity prices. [18]

This result is consistent with the views of economic policymakers at the time. In 2005, before the full scale of the terms of trade boom was apparent, then RBA Governor Ian Macfarlane suggested that the economy was likely to experience a period of slower growth with, ‘GDP growth rates starting with the numbers 2 or 3 rather than 3 or 4 for a time’ (Macfarlane 2005, p 8). [19]