ECB targets and imbalances

AmazingExpert

chapter-11-extending-the-open-economy-model--oil-shocks--imbalances_1.pdf

Home >Business & Finance homework help >Economics homework help >ECB targets and imbalances

Chapter 11

Extending the open economymodel: oil shocks and imbalances

[This is a draft chapter c� Wendy Carlin & David Soskice, 2013]1

11.1 Overview

In this chapter, we extend the open economy model in a way that helps us to discuss some interesting puzzles. The puzzles relate to how economies were op- erating in the fifteen years that preceded the global financial crisis. This was the era of the Great Moderation. High and volatile inflation appeared to have been eliminated from the advanced economies and there were dramatic reductions in unemployment in many European economies, where it had remained stubbornly high in the 1980s. Whereas the oil price shocks of the 1970s had caused ma- jor economic disruption, oil shocks in the 2000s were absorbed relatively easily. Policy makers and some academic observers were confident that a decade of labour market reforms and a better policy-making environment, exemplified by the kind of inflation-targeting regime discussed in Chapters 4 and 9, had played a part in producing the improvements in performance. Yet, the unfolding of the global financial crisis and the Eurozone sovereign

debt crisis from 2008 revealed that beneath the surface of the Great Modera- tion, imbalances were building and with them, the pre-conditions for crisis. In Chapters 7 and 8 we focused on the build-up of a leverage cycle in a number of advanced economies. In this chapter, we introduce the tools that are useful in understanding the international dimensions of these crises. We extend the basic macroeconomic framework to allow us to analyze these issues.

1Acknowledgements: David Hope has provided excellent research assistance for this chap- ter. We are very grateful to Liam Graham and Javier Lozano for comments on the chapter.

2CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALANC

First, we show how shocks to commodity prices, such as oil prices, can be analyzed in the AD − BT − ERU model. This analysis is motivated by the severe disturbance to the global economy caused by the oil shocks of the 1970s. The two oil shocks in 1973 and 1979 were followed by years of low growth, high inflation and rising unemployment – so-called ‘stagflation’. It is striking that when the oil price increased sharply again in the 2000s, the macroeconomic effects were very different. In spite of near-record oil prices (in real terms), inflation remained subdued and unemployment in many countries was at its lowest level for several decades. This raises the question as to why the outcome was different. We shall see that the extended model helps provide insight on this. The second extension to the open economy model in this chapter centres on

imbalances and the interdependence of countries. Some countries run current account surpluses; others current account deficits. We need to have a way of thinking about whether these international payments patterns are problematic or whether they simply reflect benign differences among countries, for example, in investment opportunities or natural resource endowments. Some observers argue that the very large build-up of global current account

imbalances in the years before the financial crisis played a causal role in the crisis. But before tackling that issue, we need to extend the AD − BT − ERU and 3-equation models to the two-bloc case. The simplest way of thinking about the interaction between economies is to assume the world is made up of just two large blocs of economies. We assume that in each bloc there is an inflation- targeting central bank. Using the extended model, we can explain how it was possible for inflation targeting central banks in the two blocs to successfully keep inflation close to target and yet for there to be persistently rising external imbalances.

11.1.1 Howdoes a commodity price rise affect themacro- economy?

As far as an oil-importing country is concerned, we shall see that an oil shock is a combination of two different shocks: it is a negative external trade shock (which is also a negative aggregate demand shock) and a negative external supply shock. However, in the contemporaneous analysis of the first oil shock in 1973, policy makers concentrated on the first feature of the shock. We provide a more detailed description of the oil crises in Section 11.2.1 below. It was the depressing effect on aggregate demand, employment and the trade balance that preoccupied policy makers. Let us take each aspect of the shock in turn.

Why does an oil shock depress aggregate demand? Oil is a key im- ported input to production and also an important element in the household consumption bundle. When the oil price rises, this depresses aggregate demand in oil importing countries. Thinking of the definition of aggregate demand, the higher oil price reduces net exports by raising the real import bill. From the

11.1. OVERVIEW 3

Initial real exchange rate

Initial equilibrium output

ERU

Output

Real exchange rate

New equilibrium

output

AD′

BT ′

New real exchange rate

Figure 11.1: The external trade effects of a rise in the price of oil

perspective of households, when the oil price goes up, the price of petrol at the pump rises. Firms will also pass on the higher price of energy in the prices of their goods. These price rises reduce the real incomes of households and unless they can borrow to smooth this shock, consumption will fall. This reduction in aggregate demand (because of the fall in (X − M) as the real cost of imported oil rises) shifts the AD curve to the left in the AD − BT − ERU model. The fall in net exports (because of the higher import bill) also depresses the

trade balance and we can show this in the model as a leftward shift in the BT curve. The shift in the AD and in the BT curves are shown in Fig. 11.1. As we saw in Chapter 10 (Fig. 10.11), it is not a coincidence that the new BT curve intersects the new AD curve vertically above the initial equilibrium. If we do the mental experiment of thinking about a real depreciation that would fully offset the effect on aggregate demand of the oil price shock and leave output unchanged at its initial level, this gives a point B on the new AD curve (AD′). Since the way this experiment works is by improving competitiveness and restoring net exports to their initial level, it must be the case that there is also trade balance at this point. From Fig. 11.1, we can see that the external trade shock would mean a new

medium—run equilibrium at point C, with higher unemployment and a trade deficit. It is these aggregate demand and trade balance aspects of the shock that were at the top of policy makers’ minds in 1973. They focused on trying to offset the implications of the shock for aggregate demand and the external balance.

Why is an oil shock a negative supply shock? It turns out that an oil shock is also a negative supply shock: This means not only that it raises unemployment as a consequence of the fall in aggregate demand, but that it also shifts the ERU curve to the left. Why is this? Think back to a household

4CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALANC

curvePS

curvePS ′

Initial real exchange rate

Initial equilibrium output

curveWS

ERU

Output

Real exchange rate

Initial equilibrium employment

New equilibrium

employment

Employment

New equilibrium

output

ERU ′

Real consumption wage

Initial real wage

New real wage

Figure 11.2: A negative external supply shock: an increase in the oil price shifts the ERU to the curve left

faced with a higher cost of living as a consequence of the energy price price. In Chapter 10, we discussed the fact that households are likely to view their welfare in terms of their real consumption wage, i.e. they will evaluate the value of their nominal wage according to the consumption bundle they can buy with it. If the real value of their nominal wage is reduced by higher energy prices, this shifts the price-setting curve downwards (for a given real exchange rate), opening up a gap between the real consumption wage they expect at a given rate of unemployment and the real value of the wage they get. In Fig. 11.2, we can see that this downward shift in the P S curve has the effect of shifting the ERU curve to the left. Thinking about an oil price increase as a negative supply shock helps to

explain why the oil shock in 1973 led not just to a one-off increase in inflation, but to rising inflation – at a time when unemployment had also risen. There will be upward pressure on inflation as long as there is a gap between the W S curve and the new lower P S curve: unemployment must increase if the economy is to achieve stable inflation. In other words, if the policy maker tries to offset the effect of the fall in aggregate demand by monetary or fiscal stimulus, the economy will experience rising inflation.

For example, suppose the policy maker used a fiscal stimulus to shift the aggregate demand curve back to its initial pre-shock position (from AD to AD′

11.1. OVERVIEW 5

Initial real exchange rate

Initial equilibrium output

ERU

Output

Real exchange rate

New equilibrium

output

AD′

BT ′

New real exchange rate

ERU ′

Figure 11.3: An expansion of government spending in response to a rise in the price of oil

in Fig.11.3). Would this offset to the demand effects of the shock restore equilib- rium in the economy? The answer is ‘no’ if the relevant post—shock supply-side equilibrium is on ERU′: the new equilibrium unemployment is higher as shown by point B in Fig. 11.3. Since at A the real wage has been pushed down by higher energy prices, wage setters will get higher money wages to compensate them and inflation will no longer be constant (i.e. there is rising inflation at A). This is reflected in the diagram by the fact that point A is to the right of the relevant ERU curve labelled ERU′.

What is the role ofmonetary policy? In the analysis so far, the economy ends up at point B with constant inflation. Unemployment is higher than it was initially and there is a substantial deterioration in the economy’s external balance (note the position of BT ′). Although it is not shown explicitly in the diagram, it is also the case that the government’s fiscal balance will have deteriorated due to the use of expansionary fiscal policy in boosting aggregate demand, as well as because of the operation of the automatic stabilizers. If the government wants to restore output and unemployment to their pre-oil shock levels, then it may try to use expansionary monetary policy. Outside an inflation—targeting regime, monetary policy can be used to stim-

ulate aggregate demand by producing an exchange rate depreciation through a reduction in the interest rate. In a fixed exchange rate regime, the same result could be achieved by a devaluation. This would take the economy toward point A – the economy’s starting point. However, we can see immediately that this will provoke a rise in inflation.

Why? Because at point A, the economy is to the right of the new ERU′ curve: the depreciation will cut real wages and lead to compensation through higher nominal wage settlements. Higher wage and price inflation will result. As

6CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALANC

the impact of higher inflation on home’s competitiveness kicks in, the economy will move to the south—west toward point B. To offset set this recession, the government would have to implement another loosening of monetary policy. Another burst of inflation would ensue. (Although the cause is different, this is the same kind of inflationary process as described in Fig. 10.14 in Chapter 10).

How could a supply-side policy help? Although many countries experi- enced the combination of high inflation and unemployment in the 1970s and 1980s, this was not true of all – see Fig. 15.1 in Chapter 15 for a comparison of advanced economy unemployment rates during this period. We can use the model to help provide a possible explanation. If the negative supply-side ef- fect of the shock could be offset by an appropriate supply-side policy, then the stagflationary consequences of the oil shock would be mitigated. One such policy is a wage accord. A wage accord, or incomes policy as it

was sometimes called at the time, is typically a tripartite agreement between the government, employee unions and employer associations. The accord means that employees accept wage restraint in order to maintain higher employment than would be the case otherwise. These accords have the effect of shifting the wage-setting curve downward and hence offsetting (at least partially) the downward shift of the price-setting curve. If the wage accord fully offsets the effectof theoil shock, then the ERU curvewill not shift to the left. Such apolicy would reduce both the rise in unemployment and the rise in inflation associated with the oil shock. The downside of these policies is that they are hard to sell politically, as coupling rising oil prices with wage restraint means workers have to accept a reduction in their real incomes. The discussion in Chapter 15 about differences in institutional structures across countries helps to explain why the oil shocks of the 1970s resulted in stagflation in some countries but not others.

Whydid the oil price hike in the 2000s not lead to stagflation? Over the course of the period between 2002 and 2008, real oil prices almost doubled, taking them to a level above that of 1980. Yet, inflation remained low and unemployment was falling in many countries. An important reason that the depressive demand effects were muted in the 2000s was the very different be- haviour of financial institutions. Banks were very keen to increase their lending and in the US for example, they allowed households to withdraw equity from their house to enable them to maintain their consumption in the face of the oil price increases. The behaviour of banks in this period is discussed at greater length in Chapter 8.

On the supply side, it seems that workers were less able or less inclined to get compensation through their wages for the higher oil prices in the 2000s than had been the case in the 1970s. This may be related to the decline in the role of unions in wage setting over this period (see Fig. 3.5 in Chapter 3). If this was the case, then in terms of the model, both the aggregate demand curve and the ERU curves would have shifted less to the left than was the case in the 1970s. Finally, with their inflation-targeting mandates, central banks were not inclined

11.1. OVERVIEW 7

to try to offset the effect on demand by loosening monetary policy in the 2000s.

11.1.2 Interpretinganeconomy’s sectorfinancialbalances: does a current account imbalancematter?

To answer this question, it is useful to explain circumstances under which a current account imbalance is benign. The logic is familiar from the discussion of household decisions in Chapter 2. At different stages in their lifecycle house- holds would ideally like to borrow or lend in order to maintain a fairly smooth consumption path. A phase of borrowing would be expected when income is below its expected longer—term level; a phase of lending would be expected dur- ing years of higher than ‘permanent’ income. The ability to borrow and lend, i.e. access to the capital market, allows the household to improve its welfare relative to a situation in which its consumption is tied to its current income.

The same logic can be applied to a country. A good example is a situation in which a countrygets a windfall increase in its wealth and hence in its ‘permanent income’ because of the discovery of a natural resource. Applying the same logic as in the household case, the country’s permanent income has gone up, which means consumption can be higher now and into the indefinite future. If the country has access to borrowing on the international capital market, then current consumption of home residents can go up immediately on the discovery of the natural resource before any of it has been extracted. In practical terms, this means a rise in imports as home residents purchase more goods from abroad to sustain their higher consumption.

If we assume the initial position was of current account and trade balance, we would observe a deterioration in the current account on discovery of the nat- ural resource. In this example, the current account deficit is not a signal of any weakness in the performance of the economy: indeed, the current account deficit reflects the increase in long-run wealth of the country, which allows residents to improve their living standards immediately because they can borrow on inter- national capital markets to fund the higher level of imports. Once the revenue from the natural resource comes on-stream, the accumulated debt associated with the years of current account deficits can be repaid.

However, just because we can think through an example in which a current account deficit is benign does not mean this is always the case. Let us take a very different example. Suppose the economy is characterized by a property price bubble. This can lead to a consumption boom as households feel wealthier as a consequence of rising house prices. For the economy as a whole, higher con- sumption leads to higher imports and to a deterioration in the current account. The country is accumulating debt to the rest of the world for as long as the current account deficit persists. Unlike the resource windfall case in which the means for repaying the debt are ‘in the ground’, the house-price boom does not create more wealth that can be used to repay the debt as it comes due. When the bubble bursts and house prices fall, the illusion of higher permanent income for the country is shattered and the country will have to find a way of servicing

8CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALANC

��

�

� ��

��

�

��

� � ��

� �� !�

"�� #

$��

�%&��

&� �

��

� ��

Figure 11.4: Current account balances - Countries with large surpluses or deficits: 1980 - 2010 Source: IMF World Economic Outlook, September 2011

the higher debt it has accumulated. This will normally entail a fall in living standards.

11.1.3 Inflation targeting in a two-bloc world

One of the striking features of the global economy from the year 2000 in the late phase of the Great Moderation was the mounting external imbalances of a number of large economies. Fig. 11.4 shows that the US, Spain and the UK had rising external deficits of a globally significant size. The counterpart to these deficits were the growing surpluses of China, the oil-producing countries and Germany. Japan’s substantial surplus remained fairly stable over the 2000s. In the same period, central banks in both developed and emerging and de-

veloping economies practiced inflation targeting and were seemingly successful at achieving low and stable inflation. Fig. 11.5 shows how inflation in these two country blocs remained low in the 2000s. This is in contrast to previous periods, which were blighted by high and volatile inflation, particularly in emerging and developing economies. In hindsight, the imbalances that emerged in the 2000s seem unhealthy, a

sure sign of the troubles that were to come. In the wake of the financial crisis,

11.1. OVERVIEW 9

�

��

�

��

� � � �

��

� ��

��

� � � � � ��

��

Figure 11.5: Consumer price inflation, annual average percentage change: 1970 - 2010 Source: IMF World Economic Outlook, September 2011

politicians and economists have been quick to extol the virtues of "rebalancing" – i.e. countries moving towards more balanced current account positions. We can use a 2-bloc version of the 3-equation model to explore interdependence among countries and to show how the response of an inflation-targeting central bank in one bloc to a bloc-specific shock affects the economy and the policy maker in the other bloc. We show that both blocs can achieve their inflation target, but current account and real exchange rate divergences emerge if there are different shocks and/or patterns of demand in the two blocs. The 2-bloc model will be set out and used to analyze the dynamic adjustment of two blocs to economic shocks in Section 11.2.5.

11.1.4 Differentmedium—run‘growth’ strategiescancause global imbalances

One way of understanding the global imbalances that arose in the pre-crisis pe- riod is to see them as the outcome of the choice of different medium—run ‘growth’ strategies by significant global economies. For example, the US economy expe- rienced consumption and housing booms and rising government spending under George W. Bush. From the analysis in Chapter 10, we know that a constant inflation equilibrium can be consistent with external imbalance in a small open economy. For example, we showed there that a government in a small open economy (with a downward-sloping ERU curve):

• can encourage higher domestic demand (public or private),

10CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

World real interest rate

OutputWorld output

Stabilizing real interest rate

World

Real exchange rate

OutputNew equilibrium output

ERU

AD BTInitial real

exchange rate

Real exchange rate

Output

Initial real exchange rate

Bloc A Bloc B

AD′New real exchange rate

Initial equilibrium

output

AD BT

Initial equilibrium output

AD′ BNew real

exchange rate

New equilibrium

output

ERU

Figure11.6: Differentgrowthstrategies: oneblocpursuingexpansionarypolicies and the other pursuing restrictive policies

• which will be associated with lower unemployment and

• higher real wages (via an appreciated real exchange rate), which is

• consistent with stable inflation.

If this were to improve a government’s prospects for re-election, then why don’t all countries act this way? If they all tried to do so at the same time, then since the world as a whole is a closed economy, this would be modelled by a rightward shift of the world IS curve in a closed economy model and it would not be possible to maintain constant inflation. In Fig. 11.6, we provide a simple depiction of the medium-run equilibrium

in the global economy in the pre-crisis years. Bloc A represents those economies who adopted a policy of supporting growth through domestic demand, such as the US, UK and Spain. At the level of the global economy, these expansionary policies were, however, offset by the bloc B economies, who depressed domestic

11.1. OVERVIEW 11

demand in order to keep their real exchange rate depreciated and support their export-led ‘growth’ strategy. Two important countries where the attention of policy-makers was on restraining domestic demand and promoting net exports in the 2000s were Germany and China. In the stylized example in Fig. 11.6, bloc A is in a constant inflation equi-

librium with lower unemployment and an external deficit and bloc B is in a constant inflation equilibrium with higher unemployment and a trade surplus. Assuming that the demand shifts are opposite and symmetric, the world econ- omy remains at the unique constant inflation equilibrium as shown in the upper panel of the Fig. 11.6. In the 2000s, the pattern of current account trends shows that although

some countries were choosing the buoyant domestic demand strategy, others were doing the opposite. Using Fig. 11.6, we can view the global economy as characterized by stable inflation, because one set of countries adopted policies of buoyant domestic demand with persistent external deficits, whilst the other set adopted export-led growth strategies with external surpluses. It is interesting to understand why not all countries had the same incentive to encourage higher domestic demand. We can distinguish between an export-oriented growth strategy of China

and Germany and a finance-oriented growth strategy of the US, UK and Spain. These strategies resulted in the emergence of large external imbalances devel- oping from the beginning of the 2000s, with export surpluses rising sharply in China and Germany, and deficits rising rapidly in the US, and to a lesser extent in the UK and Spain (see Fig. 11.4). As we discussed in Chapters 7 and 8, financial deregulation in the USA

began in the early 1980s and culminated in 1999 with the repeal of the Glass Steagall act. In the UK, it was the ‘big bang’ of 1986 that signalled the start of rapid financial deregulation. In the US, this led to the extension of credit to low income households for mortgages. Financial deregulation fuelled housing booms in many other countries, including Spain, where the creation of the Eu- rozone also affected access to and the cost of loans. In many of the economies with finance-orientated growth strategies, household savings ratios also fell to historically low levels in the 2000s. However, not all countries were characterized by the same pattern. In par-

ticular, in terms of countries with global impact, China and Germany looked very different. Both countries had rapidly increasing current account surpluses. Looking first at China, in spite of a very high and rising investment rate (as a per cent of GDP) and rapid growth at rates close to 10 per cent per annum, sav- ings were even higher (reflecting increasing savings by firms offsetting a falling savings rate of households). The Chinese government favoured an export—led growth strategy to create a large globally competitive manufacturing sector. It was prepared to prioritize this over a more balanced growth pattern, which would have allowed the exchange rate to appreciate and real wages and domestic consumption to rise. As Germany emerged from its periodof financing reconstruction in East Ger-

many following reunification in 1990, government policy concentrated on setting

12CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

policy so as to encourage the restoration of competitiveness in the export sec- tor, which had been eroded during the post—unification boom in the early 1990s. The government kept fiscal policy tight and supply—side reforms were focused on increasing the cost of job loss: both had the effect of encouraging wage— restraint. Against the background of weak demand at home, German firms sought to take advantage of the opportunities available to reorganize produc- tion networks in Central and Eastern Europe and to sell to the rapidly growing markets in China and elsewhere in the emerging economies. Real wage growth was low, consumption depressed and the household savings rate increased. Al- though China was growing very fast and Germany slowly, the common factor of an export-orientated growth strategy meant that in both countries national savings were higher than investment and there was an export surplus. In fact, the German preoccupation with export-led growth is not a recent

development, as financial journalist John Plender points out, Germany has been characterised by a trade-orientated growth strategy and a mistrust of high fi- nance for centuries:2

"Mercantilism and the fear of sophisticated finance have historically gone hand in hand in Germany and other parts of northern Europe. In the 15th century the cities of the Hanseatic League were pro- foundly suspicious of credit. They largely excluded foreign bankers. Merchants tried to balance trade bilaterally, relying partly on barter while making some use of coin. The economic historian Raymond de Roover reckoned the League’s credit institutions were about two centuries behind the Italians in 1500."

The third important source of rising current account surpluses was the im- pact of the dynamism of the emerging economies on oil prices, and hence on the export surpluses of the oil-producing countries.

By the mid-2000s, the magnitude of the surpluses relative to the global economy was unprecedented. Those surpluses were recycled to other regions. This echoed on a global scale the recycling of the surpluses of the oil exporters following the OPEC shocks of the 1970s to Latin America, which had created the basis for the subsequent regional debt crisis.

In a very stylized way, we can relate this discussion to Fig. 11.6. The figure is a depiction of medium-run equilibrium in the global economy in the pre-crisis period, with bloc A representing the finance-orientated countries (e.g. US and UK) and bloc B representing the export-focused countries (e.g. China and Ger- many). The boom in demand in bloc A caused an appreciation in the exchange rate and a current account deficit, whereas the opposite macroeconomic policies in bloc B ensured a depreciated exchange rate and a trade surplus. The strate- gies of the two blocs offset each other, which meant that world interest rates could be kept low during the pre-crisis period (as shown by the upper panel of Fig. 11.6) without causing either bloc to miss their inflation target.

2See Plender (2011).

11.2. MODELLING 13

11.2 Modelling

11.2.1 Oil shocks

In this section, we show how to include the role of imported raw materials in the macro model. As we have seen the most important real—world application of this for the advanced economies is to the oil shocks in the 1970s and 2000s. However, the insights from modelling shifts in the terms of trade faced by economies are much more broadly applicable.

External supply shocks (e.g. oil shocks) in the AD-BT-ERU model

An external supply shock is defined as an unanticipated change in the world terms of trade between manufactures and raw materials: a change in the world price of oil is a good example. This type of shock combines the effects of an external trade shock (as seen in the previous chapter) with a supply-side impact on the price-setting real wage curve. The consequence is that there is a shift in the AD curve, in the BT curve, and in the ERU curve, with all three curves shifting in the same direction.

To see why the ERU curve shifts, we need to look closely at what is meant by a change in the world price of oil. If we say that the world price of oil rises, this means that it rises relative to the world price of manufactured goods, where τ = P ∗rm/P

∗

mf . In other words, we are talking about a change in relative prices, or to put it another way, a change in the real price of oil. The price-setting

curve is defined for a given real exchange rate, Q ≡ P ∗

m f e

Pm f , where we now specify

that this is the relative price of manufactured goods: oil is excluded.

Now suppose that the world price of oil rises (↑ τ). For a given real exchange rate (q0 in Fig. 11.7), a rise in the price of an essential input like oil raises costs for firms in the home economy. These costs are passed on in higher consumer prices, which reduces the real consumption wage of workers. If firms are to protect their profit margins in the face of the oil price rise, then real wages must be lower. Hence the price-setting real wage curve shifts downward when the world price of oil rises (see Fig. 11.7). This implies a leftward shift in the ERU curve.

A very simple way to model the impact of oil is to assume all imports are oil and that the consumer price index is the price index of home value-added (i.e. marked—up unit labour costs) plus the unit cost of imported oil. In this simple setting, petrol at the pump is a ‘manufactured good’. Then

Pc = Pmf + vτP ∗

mfe

where Pmf = W

(1−µ)λ is the price index of home value added and v is unit

materials requirement. This implies a price-setting real wage:

wPS = (1− µ)λ

1+ vτ Q .

14CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

( )0 0,PS q τ ( )0 1,PS q τ

ey ' ey

ERU

��

eN ' eN N

ERU ′

Figure 11.7: A negative external supply shock: an increase in the oil price (↑ τ) shifts the ERU to the curve left

Any rise in τ reduces the price-setting real wage. Note that any fall in unit materials requirement through increased energy efficiency, for example, would tend to offset this. We can now analyse the full impact of an exogenous and permanent change

in the world price of an essential commodity such as oil. We take the case of a rise in the price of oil. For simplicity, we assume that the home country only imports oil – it does not import final goods. This changes nothing essential and allows for a more direct examination of the issue at hand. We can investigate the three effects:

(1) the impact on aggregate demand

(2) the impact on the trade balance

(3) the impact on price and wage setting and hence on the ERU curve.

The aggregate demand and trade balance effects of an oil shock are modelled in the same way as an external trade shock. In this case, there is a downward shock to net exports because the increase in the cost of the essential imported raw material absorbs a higher proportion of home income at a given real ex- change rate. This shifts the AD curve and the BT curve to the left. We write

11.2. MODELLING 15

out the trade balance equation from Chapter 10 to include the external terms of trade, τ.

Since PM = P ∗

mf e when imports are manufactured goods, PM = P ∗

P∗ m f

P∗mf e

when imports are raw materials. Hence, imports in real terms, M, are:

M = τ Q × M(τ Q, y)

Holding all other variables constant, a rise in τ increases the import bill, de- pressing net export demand;

X − M = X(Q, y∗)− τ Q × M(τQ, y).

This reduces X − M, which shifts the BT and the AD curves to the left. As explained in Fig. 10.11 in Chapter 10, the new AD and BT curves intersect vertically above A at A′. Figure 11.8 summarizes the effect on all three curves of an oil shock. Given

the shift in the ERU curve, if the government tries to restore aggregate demand and output to its pre—shock level at point A, then the inflationary consequences of the commodity price rise are clear. Following the external supply shock, the initial equilibrium point A is above the new ERU curve labelled ERU′. This means that at y0, the real wage is below the wage-setting real wage because the P S curvehas shifted down due to the increase in τ. A is no longer a medium—run equilibrium. If the authorities did not attempt to offset the demand shock, the relevant

AD curve is AD′ and there would still be inflationary pressure until output had fallen to y1.

Comparing three oil shocks

This section focuses on the three major global oil shocks since the early 1970s, comparing their causes, their macroeconomic implications and the associated policy responses. Fig. 11.9 shows the path of oil prices over the last 40 years – to compare across periods, we show log real oil prices.3 This approach allows for variations in the series to be interpreted as percentage changes. The shaded grey areas show the oil price shocks that we analyse in this section. The graphs also show the movement of US consumer price inflation and unemployment over the period, providing an insight into the macroeconomic consequences of each shock. The three oil shocks we are investigating took place in 1973-1974, 1979-1980

and 2002-2008. We will go through each shock in turn, setting out its causes and its macroconomic effects. We will also investigate the dominant monetary policy response of central banks in developed economies in relation to each episode.

3This way of presenting data means that an ↑ in the index from 1 to 2 is (approximately) a doubling of real oil prices. The first step is to produce the series for oil prices in real terms, by deflating nominal oil prices by the US GDP deflator (i.e. 2005 = 100). Next, the natural logarithim of this series is taken. Lastly, the series is multiplied by 100. This method is in line with that used in Blanchard and Galí (2008).

16CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

AD′

0y1y y

q ��

ERU

BT ′

ERU ′

A′

B′

A C

Figure 11.8: The macroeconomic impacts of an oil shock

OPEC I: 1973-1974 The Yom Kippur War began in the Middle East on October the 6th 1973. This did not directly affect oil shipments, but did lead to the Organization of Petroleum Exporting Countries (OPEC) cutting production of crude oil significantly towards the end of 1973. OPEC controlled a signifi- cant proportion of world oil production at the time and this negative supply shock is thought to have been the main driver of the more than doubling of crude oil prices seen over this period – the nominal oil price rose from $4.3 to $10.1 per barrel between 1973 Q4 and 1974 Q1. The overall impact of this event is estimated to have been between a 7% and 9% reduction in world oil supply (Hamilton, 2009). In addition to this supply shock, demand pressures are thought to have played a complementary role in the oil price rise. Barsky and Killian (2002) suggest that general inflation and booming prices for other commodities contributed to the upwards pressure on oil prices.

Fig. 11.9 shows that this shock coincided with a rise in inflation to double digits. In contrast, the unemployment impact was largely felt in the year after the oil price hike, when the unemployment rate in the US nearly doubled.

In response to the first oil shock in 1973, many countries focused on the aggregate demand consequences and sought to offset them via expansionary fis- cal and monetary policies. For example, if we look at the consequences of an accommodating monetary policy, this would allow the exchange rate to depre- ciate and to offset the fall in aggregate demand. Referring to Fig. 11.8, the aim would be to keep output at y0 and move the economy from A to A

′ via a depreciation of the real exchange rate.

At point A′, however, the economy is to the right of the ERU curve. This causes inflation to rise, as workers real wage expectations are not being met (as shown in Fig. 10.8 in Chapter 10). The increase in inflation (↑ P relative to P ∗) results in an appreciation of the real exchange rate, via the equation Q = P

∗ e

P .

This lead to a movement leftwards down the AD curve, as net exports fall, until the economy is at B′, where the new AD and ERU curve intersect. This is the

11.2. MODELLING 17

��

�

� � � � � ��

� � � � ��

� � � ��

� � ��

� � � � ��

� � � ��

� � � � ��

� � � ��

� � � � ��

� � � ��

� ��

�

��

�

� � � � ��

� � � ��

� � ��

� � � � ��

� � � ��

� � � � ��

� � � ��

� � � � ��

� � ��

� � � ��

� ��

��

Figure 11.9: Log of real oil prices and US macroeconomic indicators: 1970 Q1 - 2011 Q4 Source: Federal Reserve Bank of St Louis - FRED (Data accessed January 2012)

new medium-run equilibrium (MRE), where there is no pressure on inflation to change.

The consequence of the adoption of policies like this after 1973 was the onset of so-called stagflation: rising unemployment and rising inflation (as the economy eventually adjusted from A′ to the new medium-run equilibrium at B′

with unemployment rising and a burst of inflation). Any renewed attempt by the government to keep output at its pre-shock level would prompt a further increase in inflation (as the same process outlined above repeats itself).

In the UK, government policy exacerbated the stagflation. The government negotiated incomes policies during the 1970s to compensate workers for the ris- ing cost of living. As firms sought to protect their profit margins the wage—price spiral continued. By encouraging real wage resistance (i.e. raising money wages to maintain real wages) this added to the sharp rises in actual and equilibrium unemployment and inflation seen in the UK after each of the oil shocks.4

4See Walton (2006).

18CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

OPEC II: 1979-1980 The second oil shock was once again the result of largely exogenous geopolitical events disrupting the production of crude oil in the Middle East (Hamilton, 2009). This time, production was affected by both the Iranian Revolution in late 1978 and the Iran-Iraq War of 1980. As in 1973, these developments led to sharp rises in oil prices, which affected both US in- flation and unemployment in a similar manner as the 1973-74 shock, as shown by Fig. 11.9. Again, the unemployment response lagged the inflation response, with unemployment not peaking until 1982, at a rate of over 10%.

When the 1979-80 oil shock struck, the nature of the shock was better un- derstood and many countries attempted to use tight monetary policy to prevent exchange rate depreciation and hence prevent a big upsurge in inflation. For example, average UK official bank interest rates were 10% in 1978 Q3 and con- tinued to rise during the shock, reaching 17% in the first half 1980. From here interest rates gradually fell, but stayed above 10% until the middle of 1983. This shift of monetary policy regime to non-accommodation of inflation coincided with a change of government in the UK, as Thatcher took power in mid-1979. Her economic views were highly influenced by Milton Friedman’s "Monetarist" school of thought and this was reflected in her decision to squeeze inflation out of the system through tight monetary policy (even if it meant unemployment would be above equilibrium for some time). Thatcher’s Monetarist experiment is discussed in detail in Chapter 13.

In terms ofFig. 11.8, adoption of tight monetarypolicywould mean that out- put would fall from A to B′, but without the hike in inflation that accompanied loose monetary policy. The high rates of inflation and the sluggish adaptation of wage and price setters to tighter monetary policy meant that most countries experienced years of high unemployment before inflation was reduced to low levels in the 1990s. This shows that governments were generally unsuccessful in introducing the types of supply-side reforms (e.g. wage accords) discussed in Section 11.1.1, which would have mitigated the inflation and output responses of the oil shocks.

Oil price rises 2002-2008 There was a persistent and marked rise in oil prices between 2002 and 2008, which culminated in a peak real oil price in 2008 Q2 that exceeded that reached in the 1979-80 oil shock.5 The forces driving this movement in prices were distinctly different from the episodes in the 1970s, with demand factors in the global oil market playing a much larger role. The 2000s was a period of rapid economic expansion for emerging markets and especially China, whose demand for oil increased exceptionally fast over this period. In addition to this, world oil production stagnated between 2005 and 2007, partially driven by a decline in Saudi Arabian production. These demand and supply factors produced upwards pressure on prices, as more nations were actively competing for a finite amount of resources.

5There were some brief periods between 2002 Q1 and 2008 Q2 where oil prices fell, but they were small and quickly reversed (see Fig. 11.9). In light of this, we threat the whole period as one oil shock.

11.2. MODELLING 19

A further component of the third oil shock that did not feature in the oil shocksof the1970s was speculation in financialmarkets. Hamilton(2009) argues that the excessively high oil prices reached in 2008 were in part influenced by the flow of dollars into commodity futures contracts. Thepath of themacroeconomy during this pronounced and consistent rise in oil prices was unexpected given past experiences. Instead of stagflation, the third oil shock coincided with a period of falling unemployment and low inflation (for the US, see Fig. 11.9). What factors canaccount for theverydifferentmacroeconomicconsequences?

Can the model help explain why things were different this time around? On the demand side, two factors appear to have been important. Firstly, greater access to credit meant that households were able to cushion themselves against the in- crease in energy costs by withdrawing equity from their houses (Feldstein, 2006). Secondly, there is some evidence that substitution away from energy-intensive activities was easier. On the supply side, both labour market reforms and inflation-targeting

macroeconomic frameworks appear to have made wage setters less inclined or able to secure compensation in their wages for higher imported energy costs. This would be reflected in a smaller leftward shift of the ERU curve. The im- portance of labour market reform was emphasized by David Walton, a member of the Bank of England’s MPC in a 2006 speech, when he cited the increased flexibility of the UK labour market as a key reason for the oil price rises of the first half of the 2000s not disturbing the UK’s low unemployment equilibrium.6

Central banks in the developed world have learnt through experience of the dangers of oil shocks for the macroeconomy and in particular for inflation. The change in policy stance since the early 1970s and the increased emphasis on keeping inflation expectations firmly anchored in the face of oil shocks was displayed in a 2004 speech by Edward M. Gramlich, then a member of the Board of Governor of the Federal Reserve,7

"I must stress that the worst possible outcome [of an oil shock] is not these temporary increases in inflation and unemployment. The worst possible outcome is for monetary policy makers to let inflation come loose from its moorings."

11.2.2 Sector financial balances

Do current account imbalances matter? An intertemporal approach to the balance of payments

Does it matter from an economic perspective if there is a current account or trade deficit or surplus in the economy? To answer this question, it is necessary to recall that any non-zero current account reflects a change in the country’s wealth. If the home country has a current account surplus then this means that it is lending abroad – if it has a current account deficit, then it is borrowing

6See Walton (2006). 7See Gramlich (2004).

20CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

from abroad. Since this borrowing will have to be repaid (with interest) in the future, the trade deficit represents a decline in the home country’s wealth. A trade deficit will implya current account deficit unless the homecountry receives a sufficient net inflow of interest and profit receipts on the foreign assets that it owns. A decline in wealth sounds like a bad thing – but this is not necessarily the

case. When a student goes into debt to finance their university studies, their financial wealth falls. The wisdom of this move depends on the extent to which the university education increases the student’s human capital and improves their earning capacity. In the same vein, we can show how countries can rationally use borrowing

from abroad to smooth consumption when there is an expectation that future income will exceed current income. Starting from the accounting identities, we can show the link between the current account and consumption.

X − M ≡ y − C − I − G

CA ≡ X − M + INT

≡ y + INT − C − I − G = �y − C,where �y = y + INT − I − G.

By defining �y = y +INT −I −G i.e. GDP plus net interest from abroad minus investment and government spending, we can see that the current account can be viewed as �y − C, or savings. We shall call �y aggregate household income and note that it includes net interest from abroad. The intertemporal model of consumption of Chapter 2 focuses on the objective of consumption smoothing. Using the identity, we can see that in the open economy, fluctuations in the current account can allow aggregate consumption to remain constant in the face of fluctuations in income. The intertemporal model of the current account (ICA) represents the CA

as a forward-looking function of income and asset returns. And the cumulated value of past CA balances is defined as the net foreign asset (NFA) position. This model of the current account requires two key assumptions:

• Perfect international capital mobility: Home residents can buy or sell for- eign bonds with the fixed world interest rate, r∗, in unlimited quantities at low transactions costs.

• Domestic consumption is set by the infinite horizon rational expectations permanent income hypothesis. As discussed in Chapter 2, this amounts to perfect consumption smoothing in expectation when the real interest rate (r) is equal to the subjective discount factor (ρ).

Using the notation introduced in Chapter 2, we can write the ICA model as follows:

CAt = − ∞�

i=1

( 1

1+ r∗ )iEt∆�yt+i,

11.2. MODELLING 21

where, CAt is the current account balance at time t, ∞�

i=1

is the sum from period

t + 1 to infinity, r∗ is the world interest rate, Et is expectations at time t and �yt+i is aggregate household net income at time t+i. In this framework, just like a household, a country has an intertemporal budget constraint. Consumption smoothing by borrowing and lending makes sense for individuals and the same can be said for countries. If a country experiences a temporary decrease in incomethis period(i.e. fall in �yt due forexample toanexogenous fall inexports), then there is an expectation that future income will be higher than current income (i.e. Et∆�yt+i > 0, when i > 0). In this scenario, a current account deficit in period t would simply reflect

optimal borrowing at the world interest rate, r∗, to smooth consumption against the adverse income shock. This borrowing from abroad would need to be repaid in the future, so from period t + 1 onwards, the country would run a series of small current account surpluses as the debt is repaid (and the NFA gradually returns to zero). This illustrates that a country’s current account deficit can be a reflection of its ‘permanent income’ and therefore that the possibility of bor- rowing from abroad is a rational method of increasing the utility of its citizens. This example is analogous to a household borrowing to smooth their income when they have a bad income shock or when their expected lifetime income increases, as was discussed in detail in Chapter 2. Consider the example of a country that discovers a natural resource, such as

oil or diamonds (see Fig. 11.10). This raises the country’s wealth and therefore its permanent income (i.e. Et∆�yt+i > 0, when i > 0). The extraction of the resource commonly takes place some time after the discovery is made. However, the ability to borrow in international capital markets means the country can smooth its consumption, raising living standards even ahead of the first drop of oil being extracted. There are three separate factors that contribute to the emergence of a CA deficit upon the discovery of oil in this scenario:

1. Current consumption rises due to the expectation of higher future wealth.

2. Domestic investment increases to enable the extraction of the natural re- sources. This is partly funded through international borrowing.

3. The nominal and real exchange rates appreciate, because the forex mar- ket anticipates that the real exchange rate that will balance the current account will be an appreciated one. The discovery is a positive external trade shock (as discussed in Chapter 10).

Fig. 11.10 shows the time profile of the discovery of the resource, the period for which revenues are extracted and the period after the natural resource is exhausted. Following the discovery, such a country will have a current account deficit. Its consumption is permanently higher than would have been the case in the absence of the natural resource discovery. Once the revenue from the resource extraction begins to flow the current account balance improves. The current account eventually moves to a small surplus, which continues until the

22CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

time 0

t0 t1 t2

Levels relative to those before the resource windfall

Resource windfall: discovery at t0, revenue flow begins at t1and is exhausted at t2

Consumption smoothing: consumption rises on discovery: financed by borrowing from abroad & remains constant throughout.

1st Current account is negative reflecting consumption smoothing. External debt rises. 2nd Current account improves during flow of revenue from natural resource exploitation 3rd Current account surplus repays debt built up during the first phase.

Figure 11.10: Resource windfall and the current account

country has repaid the debt built up during the initial phase. This adjustment was observed in the UK in the 1970s with the discovery of oil in the North Sea. If we think about current account imbalances more generally, for a country

that does not have very profitable investment opportunities at home, it makes sense that domestic savings are used for net investment abroad. The purchase of foreign assets that the current account surplus represents may provide a higher return than would higher investment at home. High saving economies in Asia, such as Singapore, provide examples here.

Causes and consequences of current account imbalances

• If all economic agents act rationally weighing up the relative returns from different investment opportunities, then a current account imbalance sim- ply reflects the differences in preferences, in investment opportunities and in resource windfalls across countries. This is a useful benchmark case, but a persistent current account deficit is not necessarily benign.

• For reasons ofmyopia andpoliticalpressures, a currentaccount deficitmay not reflect higher investment at home in response to especially attractive investment opportunities or because of a resource windfall – rather, it may reflect low savings because of high private consumption based, for

11.2. MODELLING 23

example, on a property price bubble or unsustainably high government consumption or it may reflect investment in wasteful projects.

A persistent or excessive current account deficit may become increasingly difficult to finance. In practice this means that under flexible exchange rates, there may be a depreciation of the exchange rate. Under fixed exchange rates, if private counterparties are not willing to finance the deficit at the world interest rate, the central bank will be obliged to sell foreign exchange reserves in order to maintain the exchange rate. In circumstances such as these, the government and central bank will be obliged to change domestic policy so as to reduce the current account deficit, for example by tightening fiscal policy.

Since a current account deficit implies a running down of the home coun- try’s wealth (an increase in its foreign liabilities or indebtedness) whereas a surplus represents an accumulation of wealth, there is an essential asymmetry between the two. Foreigners may stop lending to the home economy: they cannot stop the home economy from accumulating foreign exchange reserves or foreign assets. Or can they? If a country is believed to be ‘too competitive’, then sanctions may be imposed by its trading partners such as the imposition of quotas or tariffs. Political pressure may be applied to encourage an appreciation of the exchange rate and a shift from reliance on exports to a more balanced structure of demand. Such debates in the US centred on Japan in the 1980s and on China in the new millennium. We return to the issue of global imbalances in Section 11.2.4.

11.2.3 Sector financial balances

A major theme in the discussion of the background causes of the global financial crisis is global imbalances. What are these imbalances and how do they fit into our model? To understand what lies behind the emergence of global imbalances – current account (or trade) surpluses or deficits – it is helpful to see how the goods market equilibrium condition for one economy can be rearranged to show its sector financial balances. We can write the goods market equilibrium condition in terms of sectoral savings and investment balances.

Three sector balances are of interest: the private sector financial balance (private savings net of its investment), the government sector financial balance (taxationnetofgovernmentexpenditure), and the trade balance (net investment abroad). We assume that the economy is in a short-run equilibrium at which r = r∗. Ceteris paribus, a trade surplus means that stocks of foreign assets are increasing in the home economy. This measures the increase in the holdings of foreignwealth in thehomeeconomyand is therefore referred to as net investment abroad. We rearrange the outflows and inflows version of the goods market equilibrium condition to separate out taxation and show the sector financial

24CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

balances:8

(S − I(r∗))+(T − G) = X − M

(s1y disp − c0 − I(r

∗)) � ��

private sector financial balance

+ (ty − G) � �� government

financial balance

= X(Q, y∗)− QM(Q, y) � �� BT = net inv abroad

where ydisp is disposable income, ydisp = (1− t)y. This expression is useful because it highlights the flow equilibrium in the

economy. One sector, for example, the private sector, can only run a financial deficit if it borrows from another sector: this would mean some combination of borrowing from the government (i.e. a government budget surplus) and bor- rowing from abroad (a foreign trade deficit). Whenever the goods market is in equilibrium, private savings net of investment (the private sector’s financial bal- ance) plus the government budget surplus (the government’s financial balance) is equal to the trade surplus.

How can ‘twin deficits’ arise?

The sector financial balances also provide a useful lens with which to look at a country’s macroeconomic developments, but the equation must be used with care. It is a goods market equilibrium condition so in order to understand the implications of a shock or policy change it is first necessary to identify the shock and to work out the new goods market equilibrium.9

In order to do this, we need to use a model. In our first example, we use the AD − BT − ERU model to illustrate the twin deficits that arise following an expansionary fiscal policy. We begin in equilibrium at trade and budget balance, and in private sector financial balance. As shown in Fig. 11.11, a rise in G leads to a new constant inflation equilibrium at higher output (on the ERU curve). Both in the temporary shorter-run equilibrium at point B and in the medium- run equilibrium at C, there is a trade deficit. In both cases, output is higher than initially, which pushes up imports; in the new medium-run equilibrium, the exchange rate is appreciated, which depresses net exports (both B and C are below the BT line). What about the fiscal balance? To pin things down, let us concentrate on

the new medium-run equilibrium. Higher output generates higher tax revenues

8A step-by-step derivation is shown below: in the first line, we multiply through by y, shift imports to the right hand side and the domestic components of demand to the left hand side. In the second line, we add and subtract s1ty and in the third line, we use the fact that s1 +c1 = 1.

s1y +c1ty −c0 − I − G = X(Q,y ∗) −QM(Q,y)

s1(y − ty) + s1ty +c1ty −c0 − I − G = BT

(s1y disp

− c0 −I) + (ty −G) = BT

9A detailed discussion of the difference between an identity and an equilibrium condition was presented in Chapter 2.

11.2. MODELLING 25

*( , )AD G r r=

��

ERU

*( , )AD G r r′ =A

ey ey ′

q′

Figure 11.11: Twin deficits arising due to an expansion of government spending

but not by enough to prevent the government’s budget balance from deterio- rating.10 Thus, in the new equilibrium, the government’s financial balance has deteriorated: there is a budget deficit. Higher y also raises savings and im- proves the private sector balance (since nothing has happened to s1, c0, or I – remember that in the new equilibrium at C, r = r∗). To summarize, the increase in government spending results in a budget deficit. The sector financial balances equation highlights the fact that in the new equilibrium, this is partly financed by borrowing from the private sector (which goes into surplus) and partly by borrowing from abroad (the trade balance goes into deficit). So-called twin deficits (fiscal and trade) have emerged.

Examples Since the early 2000s, twin (or, as we shall see, triple) deficits have consistently been observed in the United States, as shown by Fig. 11.12. In the last decade, the budget deficit reflected a combination of tax cuts and expensive foreign wars. Government spending during this period was financed by borrowing from abroad: the fact that the current account deficit exceeded the government deficit indicates that the private sector was also borrowing from abroad. Hence, the triple deficit. US borrowing was the mirror image of Chinese lending (i.e. the purchase of US Treasury bills by the Chinese authorities as part of their intervention in currency markets). Canthe US current account be rationalised by using the intertemporal model

of the current account?

CAt = − ∞�

i=1

( 1

1+ r∗ )iEt∆yt+i

10The rise in G raises income, which raises savings and imports as well as tax revenue. The level of output in the new equilibrium will therefore be lower than the level that would raise taxes by the increase in G. Hence, there is a budget deficit in the new equilibrium. As an exercise, show this algebraically.

26CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

��

� ��

��

�

��

� � � �

� � � � � ��

��

Figure 11.12: United States Sector Financial Balances: 1980 - 2010 Note: The private sector balance was computed using the sector financial balances

equation (i.e. private sector balance = current account balance less government budget

deficit)

Source: OECD (accessed September 2011)

The future macroeconomic adjustment brought on by such a substantial cur- rent account deficit could be thought of as benign if the build up of debt simply reflected optimal borrowing to smooth consumption. Some economists have pointed to expectations of rapid income growth and the presence of exception- ally low interest rates (which reduces the cost of financing the government debt and encourages consumption in the present) as possible reasons why this might be the case. However, these arguments have lost some weight since the onset of the global financial crisis, which means that the macroeconomic adjustment (i.e. repaying of debt) in the United States could indeed have painful economic ramifications. This will be explored in more detail in Section 11.2.4, as will the role of global imbalances in the recent economic downturn. In our second example, we use the intertemporal model and for simplicity

assume that the exchange rate is fixed. In this case, oil is discovered. We introduce a government sector to the model discussed above and assume that the rise in the economy’s permanent income leads the government to increase its expenditureoneducationandother infrastructurebyborrowing from abroad. In this example, the private sector also goes into deficit as consumption smoothing

11.2. MODELLING 27

by households raises consumption when the discovery of oil is announced. Here there are ‘triple deficits’ in anticipation of the flow of oil. A more complicated model would include the likely effect of the oil discov-

ery on the exchange rate. Countries where there has been a natural resources windfall have typically experienced exchange rate appreciation and suffered from ‘Dutch disease’, named after the Netherlands’ experience following the discovery of natural gas in the North Sea. Dutch disease captures the idea that the non- resource tradables sector endures a loss of competitiveness as a consequence of the exchange rate appreciation. The exchange rate appreciated when gas was discovered because the foreign exchange market anticipated that the real exchange rate at which there would be current account balance was an appreci- ated one (as the AD-BT-ERU model predicts when there is a positive external trade shock).

11.2.4 Global interdependence& imbalances

As Fig. 11.4 showed, a very striking characteristic of the period of the 2000s – the late phase of the Great Moderation (as discussed in the introduction to this chapter)–was themounting external imbalances of large countries in theglobal economy. The US, Spain and the UK had rising external deficits of a globally significant size. The counterpart to these deficits were the growing surpluses of China, the oil-producing countries and Germany. Japan’s substantial surplus remained fairly stable in the 2000s. In the same period, central banks in both developed and emerging and developing economies practiced inflation-targeting and were seemingly successful at achieving low and stable inflation. Fig. 11.5 in Section 11.1 shows how inflation in these two country blocs remained low in the 2000s. This is in contrast to previous periods, which were blighted by high and volatile inflation, particularly in emerging and developing economies. In hindsight, the imbalances that emerged in the 2000s seem unhealthy, a

sure sign of the troubles that were to come. In the wake of the financial crisis, politicians and economists have been quick to extol the virtues of "rebalanc- ing" – i.e. countries moving towards more balanced current account positions. In the years preceding the crisis economists were divided over whether the cur- rent account imbalances were benign, reflecting rational optimising by economic agents or whether they were unsustainable and posed a threat to medium-run economic stability. Caballero et al. (2006) presented a unified model which sought to explain

how large and rising US current account deficits could be coupled with low long-term interest rates and a rising share of US financial assets in world port- folios in a medium-run equilibrium. They divided the world into three regions: high-growth high-finance "U" countries (e.g. US, UK, Australia), low-growth high-finance "E" countries (e.g. Eurozone, Japan) and high-growth-low finance "R" (i.e. the rest of the world). High finance referred to a high level of financial development. The core of the model is that there is fast economic growth in the R countries (i.e. emerging markets), but that their underdeveloped finan- cial systems cannot provide enough high-quality savings instruments to satisfy

28CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

demand. This pushes up demand for savings instruments from the U and E regions. The model assumes that U has higher growth potential than E, so a disproportionate amount of global savings flow into these countries. This model explained how three macroeconomic trends that defined the early and mid-2000s – a worsening of the US current account deficit, low global interest rates and a rising share of US assets in world portfolios – could coexist. However, it has been criticised, particularly concerning the fact that the model assumptions better suit the 1990s (with the Asian financial crisis) than the mid-2000s. In the latter period, it could be argued that emerging markets did have a capacity to generate assets that others wanted (Frankel, 2006). The Caballero et al. (2006) model emphasises the importance of the US

government’s ability to create desirable financial assets. This is also highlighted in a paper by Richard Cooper on the eve of the financial crisis, which con- cluded that current account imbalances were benign, as they simply reflected rational savings decisions of individuals in economies with ageing populations (e.g. Germany, Japan, China).11 Given the more favourable demographics in the US, the argument follows that these imbalances could persist until the baby boom generation in the ageing economies reached retirement, at which point the imbalances would naturally unwind as individuals sold these assets to finance consumption during retirement. Not all economists held the view that these imbalances were sustainable. A

number thought that the global imbalances, the vast majority of which were between the US and the major surplus countries (i.e. Japan, Germany and China), would inevitably lead to a large and potentially destabilising fall (i.e. depreciation) in the dollar (Rogoff and Obstfeld, 2005; Feldstein, 2008). This would make US exports more attractive to foreigners and at the same time, make imports more expensive for US consumers. We know from Chapter 10, that if the Marshall—Lerner condition holds, a depreciation of the dollar would lead to an improvement in the trade balance. Economists at the Bank for International Settlements were some of the most

vocal proponents of the view that the successful achievement of low inflation was masking the build up of potentially hazardous imbalances. Borio and Lowe (2002) and White (2008) concentrated in particular on the accumulation of financial imbalances in a low inflation environment and how the unwinding of these could lead to a large negative macroeconomic shock. Unfortunately, their unheeded warnings proved eerily prophetic, as the global economy tumbled into financial crisis in late 2008. We will return to analyze the linkages between global imbalances and the financial crisis in the final chapter of the book.

11.2.5 A2-blocmodelwith inflation-targetingcentralbanks

The aim of this section is to set out a model which can provide a simple expla- nation of how two country blocs each with an inflation—targeting central bank, but with different patterns in demand, could produce persistent imbalances in

11See Cooper (2008).

11.2. MODELLING 29

the global economy. We set out a 2-bloc model of the world economy and show how external imbalances can be consistent with successful inflation targeting. In Section 11.1.3, we introduced the idea of medium-run global imbalances

by using the downward-sloping ERU curve. It is easier to model the dynamic adjustment of two blocs to a shock, however, if we simplify on the supply-side and revert to a vertical ERU. The downward-sloping ERU is helpful in bringing out the incentive for a country to allow a domestic demand boom to take hold and as discussed in Section 11.1.4. Here we abstract from the motivation for different patterns of demand and concentrate on how the dynamic interaction takes place between two blocs where the supply-side is identical in each and is captured by a vertical ERU (i.e. there is a unique constant inflation equilibrium in each bloc).

• Our question here is: how do inflation-targeting central banks in each bloc respond to a shock and produce successful adjustment to ensure that inflation is at target in the new medium-run equilibrium?

The role of q and r∗ in the 2-bloc model In the model, there are two blocs, A and B. To fix ideas, think of bloc A as being the deficit countries (US, UK, Spain) and bloc B as the surplus countries (China, Germany, Japan, oil exporters). In our model, the two blocs constitute the whole of the world economy. In each bloc, medium-run equilibrium is where aggregate demand is equal to equilibrium output (which is the same in each bloc at ye), with r = r̄

∗

and q = q̄. To simplify the notation, we assume that the coefficients on the interest rate

and real exchange rate are identical in each economy and equal to one. This means we can write the medium-run equilibrium for each bloc as follows:

ye = A A − r̄∗ + q̄ (bloc A)

ye = A B − r̄∗ − q̄. (bloc B)

We can see why q is positive for A and equal and opposite for B by writing out the definition of q in the 2-bloc case,

q = log

� P Be

P A

� and e =

$A $B

We assume that AA > AB (i.e. that autonomous demand is higher in bloc A). The first aspect of the model is that if we equate the right hand side of each

equation, we can simplify and get an expression for the real exchange rate, q̄:

AA − r̄∗ + q̄ = AB − r̄∗ − q̄

2q̄ = AB − AA

q̄ = AB − AA

2 < 0.

30CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

Since AA > AB, bloc A’s real exchange rate is appreciated. This is reflected in q̄ < 0. It is also clear that bloc A’s balance of trade deficit (its net imports from bloc B), which is output minus domestic absorption is negative (BTA = ye −

AA − r̄∗

= q̄ < 0) and this is equal to bloc B’s trade balance (its net

exports to bloc A), is positive: BTB = ye − AB − r̄∗

= −q̄ > 0.

It is variation of the real exchange rate between blocs that ensures that in each bloc, aggregate demand is at equilibrium and hence inflation is constant. If the blocs were symmetric in their levels of autonomous aggregate demand, q̄ = 0, which means the real exchange rate Q = 1 and common currency prices are identical in each bloc. Trade is balanced. The second aspect of the model is to note the role of the world real interest

rate: this adjusts to ensure that for the global economy (the combination of bloc A and bloc B), aggregate demand is consistent with output at its equilibrium level in the world as a whole. To see this, we write aggregate demand in block A as yD,A and similarly for bloc B. We need to ensure that

yD,A + yD,B = 2ye.

Hence, AA − r̄∗ + q̄ + AB − r̄∗ − q̄ = 2ye

AA + AB −2r̄∗ = 2ye

2r̄∗ = AA + AB −2ye

r̄∗ = AA + AB

2 − ye.

This shows that the world real interest rate adjusts to ensure a constant inflation equilibrium for the world as a whole. This happens as a result of the central banks of bloc A and bloc B adjusting the interest rate to guide inflation back to target in their blocs. Both of these central banks are forward-looking and rational and solve the model taking into account the actions of the other central bank. In the top panel of Fig. 11.13, the symmetric equilibrium is shown. In

the lower panel, we show the new medium-run equilibrium in the world and in each bloc following a permanent positive demand shock in bloc A, which pushes AA > AB. The result is a higher world real interest rate, an appreciated real exchange rate and trade deficit in bloc A and a depreciated real exchange rate and trade surplus in bloc B. Inflation is constant in the new MRE. Next we examine how the economies move from the old to the new equilibrium.

A permanent demand shock in bloc A In Chapter 9, we showed how the central bank and foreign exchange market simultaneously solved the model to work out how the central bank would choose its interest rate response to a shock and how the exchange rate would change. In the 2-bloc model, there are three parties involved in solving the model: the central banks in each bloc and the foreign exchange market. We need to use our imagination to think of these three rational actors playing a game with each other in which they have complete information about the model and about the shock.

11.2. MODELLING 31

��

�� ( ) A BA A′ >

q q−

2 ey yey yey

yeyyeyy2 ey

��

r q

ERU ERU

ERUERU

IS′

*r ′ *r

( )AAD A BT

( )BAD A BT

*( , )AAD A r BT

*( , )BAD A r ′

q′

q′−

q−

*( , )BAD A r

*( , )AAD A r′ ′

*( , )AAD A r′

( )A BA A=

Figure 11.13: The two bloc model with (a) a symmetric equilibrium and (b) a permanent positive demand shock in bloc A

To analyze how they react to a demand shock in one of them, we begin in a symmetric equilibrium with the blocs identical in every respect (the top panel of Fig. 11.13).

We assume a positive permanent shock to demand in bloc A in period zero. This pushes up output and inflation in bloc A in period zero and, just as in the closed economy, A’s central bank will have to respond to this in order to get the economy on the path back to target inflation. As we shall see, the analysis we have used for a small open economy gives good guidance as to what happens in the 2-bloc case. In response to the demand shock in period zero, the central bank in bloc A increases its interest rate to dampen activity. As usual, this is accompanied by an appreciation of the real exchange rate in bloc A and in period one, bloc A is on the MR curve and on the path back to target inflation.

Meanwhile all three actors have to consider the spillovers from the shock to bloc B’s economy. In this simple model, spillovers from A to B take place only through changes in the real exchange rate. In reality, there would also be feedback via the effect of changes in income in bloc A on imports from bloc B. In our simple model, the marginal propensity to import is zero. In a more realistic

32CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

��

*( , )AAD A r

q ERU

q−

ERU

0q q′

q′− 0q−*( , )AAD A r′ ′

*( , )AAD A r′

*( , )BAD A r ′ *( , )BAD A r

��

A B

C Z

A,B

ey ey ��

��

AD AA ′ *r *( , )AAD A r′

��

AD *r *( , )AAD A r′ ′

��

AD *( , )BAD A r ′

Figure 11.14: The two bloc model: the AD − ERU diagrams for bloc A and bloc B using the example of a permanent positive demand shock in bloc A

model, an increase in AA would not only push up output in bloc A, but also increase their demand for imports. As bloc A’s imports are bloc B’s exports, the initial increase in AA would also increase AB. We exclude this feedback mechanism from the model as it complicates the analysis. In the model, it takes one period for interest and exchange rate changes to

have an effect: hence, in period zero, nothing happens to inflation or output in bloc B. However, bloc B’s central bank observes the shock to bloc A in period zero and works out that in order to keep to its inflation target, it will need to act. Why? Because the appreciated real exchange rate in bloc A implies a depreciated real exchange rate for bloc B: whatever happens to the real exchange rate in bloc A happens (in reverse) to the real exchange rate in bloc B. Unless the central bank in bloc B raises its interest rate to offset the effect of the depreciation in its real exchange rate on aggregate demand in bloc B, there will be an increase in bloc B’s inflation in period 1. By working through the model, the three parties will figure out that the central bank in bloc B will be able to fully offset the effects of the shock on inflation in its economy. Fig. 11.14 shows the new medium-run equilibrium (as in the lower panel of

Fig. 11.13) and sketches the path of adjustment in each bloc. Fig. 11.15 shows the adjustment process in the P C − MR and IS − RX diagrams for each bloc, which can be explained period-by-period as follows:

Period 0 Both economies start at their respective bliss points - points A in Fig. 11.15. There is a positive demand shock in bloc A, which is observed

11.2. MODELLING 33

��

0IS

Tπ

eyey

ey y

*r ′ 0

IS1IS

0( ) E TPC π π=

0π 1 0( ) EPC π π=

1π

IS′

0( ) E TPC π π=

Tπ

0 Br

( )IS q′− 0( )IS q−

A,B

A,Z

1y 0y

*r ′

Figure 11.15: The two bloc model: the IS − RX and P C − MR diagrams for bloc A and bloc B using the example of a permanent positive demand shock in bloc A

by both central banks and the forex market. This moves bloc A to point B with output at y0 and inflation at π0. Bloc B has not moved from its original position. The central banks forecast their Phillips curves in the next period. In bloc A, output has changed from equilibrium, so the P C will move to P C(πE1 = π0) next period. Bloc A is not on their MR curve at point B. Their desired position on next period’s P C is point C. In order to achieve this, the central bank in bloc A raises the interest rate to rA0 , taking account of the appreciation of the exchange rate that will occur (so the UIP condition can hold). Meanwhile, in bloc B, the central bank forecasts that the P C will not move next period, as output has not deviated from equilibrium. They have however noted the actions of bloc A’s central bank will lead to a depreciated exchange rate in Bloc B (as an appreciation in bloc A is a depreciation in bloc B in this model). To counter this effect and keep the economy at its bliss point, the central bank of bloc B must raise interest rates to rB0 . This rate hike is exactly what is required to offset the boost in output that will occur as the result of the depreciated exchange rate. The new interest rates and exchange

34CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

t t

q−

π ��

0t =

0t = 0t =

0t =

Tπ

0π

0 q′

q′−

*r ′ *r

*r ′

Tπ

��

Figure 11.16: The two bloc model: the impulse response functions (IRFs) for bloc A and bloc B using the example of a permanent positive demand shock in bloc A

rates can only affect the economy with a one period. This means that bloc A ends period 0 with output at y0, inflation at π0, interest rates at rA0 and the exchange rate at q0. In contrast, bloc B remains at its bliss point, with a real interest rate of rB0 and an exchange rate of −q0.

Period 1 onwards The new interest rates and exchange rates have had time to take effect and both blocs have moved to point C. In bloc A, the higher interest rate and the appreciated real exchange rate reinforce each other and dampen aggregate demand, causing output to fall to y1 and inflation to fall to π1. In bloc B, the interest rate and exchange rate effects exactly offset each other and the economy remains at its bliss point. In bloc A, the adjustment from C to Z is very similar to the demand and supply shock cases in the open economy 3-equation model discussed in Chapter 10. The

11.2. MODELLING 35

economy adjusts along the RX curve to the new medium-run equilibrium at point Z. The RX curve has shifted upwards after the demand shock in bloc A. This is because the RX curve is pinned down by the world rate of interest, which changes as a result of the demand shock. During the adjustment from C to Z in bloc A, the central bank slowly reduce the interest rate from rA0 to r̄

∗′ to stay on their MR and RX curves. The path of interest rates takes account of the depreciation that occurs each time the interest rate is reduced. This depreciation means the UIP condition holds in all periods. In bloc B, the adjustment from C to Z is simpler. The central bank can see that the exchange rate in bloc A is going to slowly depreciate over this period, meaning that bloc B’s exchange rate is going to slowly appreciate. To offset the depressing effect from this appreciation the central bank will slowly reduce interest rates from rB0 to r̄∗′. Thoughout the entire adjustment period, bloc B does not move from its bliss point – i.e. where output is at equilibrium and inflation is at target. The adjustment to the demand shock in bloc A ends when the blocs are at point Z, where inflation is back at target and output is back at equilibrium in both blocs. The new medium-run equilibrium is however characterisedbya higher world interest rate (i.e. r̄∗′ > r̄∗), a more appreciated exchange rate in bloc A and a more depreciated exchange rate in bloc B.

There are a number of important points to highlight from this process:

• The initial interest rate hike in bloc A is greater than that in bloc B (i.e. rA0 > r

B 0 ). This is because the permanent positive demand shock takes

place in bloc A.

• Throughout the adjustment process bloc B remains at equilibrium output and target inflation (i.e. its bliss point): all that the bloc B central bank has to do each period is to adjust its interest rate to offset the effects of the real exchange rate changes arising from bloc A’s adjustment path.

• When the adjustment process is complete, both economies are at equi- librium output with inflation at target and with the same real interest rate. As shown in the bottom left hand panel of Fig. 11.13, the common (i.e. world) real interest rate is higher than in the initial medium-run equilibrium to squeeze out the higher aggregate demand. Bloc A has an appreciated real exchange rate in the new medium-run equilibrium. It also has a balance of trade deficit. Conversely, bloc B’s exchange rate is depreciated and it has a balance of trade surplus.

Lastly, Fig. 11.16 shows the impulse response functions for the key variables in each bloc. The details of the dynamic adjustment are set out in more detail in Section 11.4.1 of the Appendix.

36CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

Conclusions The model is very stylized and is based on very strong assump- tions about rationality and full information on the part of the three actors. Nevertheless, it brings out an important feature of an interdependent global economy: if economies differ in the kind of shocks they are exposed to – for example, as we saw in Chapters 7 and 8, the scale of leverage cycles was very different in different countries – it is still possible for macroeconomic equilib- rium to prevail in the sense of a constant inflation equilibrium in both blocs. The model also assumes there is no feedback from the changes in wealth that accompany the current account imbalances in each bloc. We discussed feedback of this kind in the previous chapter. The 2-bloc model illustrates how current account imbalances can arise –

when blocs are affected by different shocks and/or have different patterns of de- mand – and how they can be consistent with stable inflation. As highlighted by the intertemporal approach to the current account presented earlier in the chap- ter, such imbalances are not necessarily a problem: they may reflect an optimal response to differences in resource endowment (such as a natural resource wind- fall) or differences in preferences as countries take advantage of international capital mobility to smooth consumption.

11.3 Conclusions

This chapter has used open economy macro models to improve our understand- ing of oil shocks and global imbalances – two features of global macroeconomics that were particularly prominent in the years preceding the financial crisis. Set- ting out a frameworkwithinwhichto analyze these shocks and imbalances allows us to better understand their macroeconomic consequences and provides useful lessons for policy making. We have shown how oil price (i.e. commodity) shocks can be modelled by

introducing the world terms of trade between manufactures and raw materials. In this framework, an increase in the oil price is both a demand and a supply shock for an oil-importing economy, resulting in shifts in both the AD and ERU curves. We have also introduced two new open economy models – the intertemporal

model of the current account (ICA) and the 2-bloc model – which provide valuable insights into the sizeable current account deficits that characterised the global economy during the Great Moderation. The ICA highlights that current account can be either benign or poten-

tially dangerous, depending on whether they reflect optimizing forward-looking behaviour (e.g. borrowing to develop a newly discovered natural resource) or ir- rational myopia (e.g. a consumption boom based on an asset price bubble). The 2-bloc model can be used to show how 2 blocs in the world economy with dif- ferent patterns of demand can result in both current account and real exchange rate divergences and successful inflation targeting. This chapter allows us to shed light on some of the interesting puzzles that

characterised the years of unprecedented macroeconomic stability that preceded

11.3. CONCLUSIONS 37

the global financial crisis:

1. Why did the oil price shock of the 2000s not lead to widespread ‘stagfla- tion’, such as that experienced as a result of the 1970s oil shocks? The oil shocks of the 1970s were primarily driven by the supply-side and reflected exogenous geo-political events disrupting oil supply, whereas an increase in oil demand (particularly in fast-growing emerging markets) played a much more significant role in the 2000s shock. The latter shock coincided with a period of low inflation and falling unemployment, which was in stark contrast to the ‘stagflation’ of the 1970s. This is due to a combina- tion of factors, including (a) the increased flexibility of labour markets in the 2000s and (b) the easier access to credit and ability to substitute away from energy-intensive activities in the latter period. In addition, inflation- targeting central banks used non-accomodating monetary policy to keep inflation expectations firmly anchored in the latter period, whereas policy makers mistakenly tried to keep output at its pre-shock level in the 1970s. This failure to account for the supply-side implications of the oil shock led to significantly worse economic outcomes in the 1970s than in the 2000s.

2. What were the macroeconomic consequences of economies following dif- ferent growth strategies during the Great Moderation? There were two distinct blocs of economies during the Great Moderation; the deficit coun- tries (the US, the UK and Spain) and the surplus countries (Germany, China and the oil exporters). The former concentrated on finance and growth was fuelled by borrowing. An emphasis on the export sector was very important in both China and Germany. Although China was grow- ing rapidly, its growth was unbalanced: very high investment levels were associated with even higher saving rates. Germany grew very slowly: re- strained domestic demand and some supply—side reforms complemented restructuring of the export sector as a basis for export—led growth. The macroeconomic consequence of these different strategies was that although inflation targets were met, there was a build—up of large imbalances and interdependencies among countries.

3. Were the large current account imbalances accumulated during the pre- crisis years the result of intertemporal optimization? In the wake of the globalfinancial crisisof 2008-09, it seemsclear that thecurrentaccount im- balances were not the result of rational forward-looking behaviour. House- holds in the United States borrowed excessively and economic agents in the surplus countries (i.e. China and Germany) over-saved. This resulted in the build up of a dangerous leverage cycle (see Chapter 8), which made the global economy vulnerable to a financial crisis.

Chapters 9, 10 and 11 have primarily focused on open economy macroeco- nomics in relation to economies with flexible exchange rate regimes. In 2011, however, 19% of world output was accounted for by the Eurozone, a common

38CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

currency area with irrevocably fixed exchange rates.12 An economic model of the global economy would therefore not be complete without a proper treatment of the economics of a common currency area. The following chapter fills this gap, investigating how fixed exchange rates affect macroeconomic adjustment and stablization policy.

11.4 Appendix

11.4.1 Dynamic adjustment to a shock (optional)

In this section, we will provide the mathematics behind the dynamic adjustment to shocks in the two bloc model. We will use the example of the case where there has been a positive permanent demand shock in bloc A (but not bloc B) in period 0. The intuition and explanation for this example is contained in Section 11.2.5 of the main body of this chapter. We assume that bloc A and bloc B were in equilibrium in t = −1. So

yA −1 = A

A − r̄∗ + q̄ = ye

rA −1 = r

B −1 = r̄

∗

πA −1 = π

B −1 = π

T , and

yB −1 = A

B − r̄∗ + q̄ = ye, such that

AA = AB

and likewise for bloc B. Then in t = 0, autonomous demand in bloc A increases permanently to AA′. All 3 actors first work out the changes to equilibrium values. Bloc A and bloc B were identical before the shock, so that q̄ = 0 and r̄∗ = AA − ye = A

B − ye. After the permanent demand shock in bloc A, the new medium-run equilibrium becomes:

ye = A A′ − r̄∗′ + q̄′

ye = A B − r̄∗′ − q̄′

r̄∗′ = ye − AA′ − AB

and q̄′ = − AA′ − AB

2 .

What happens in period 0?

yA0 = ye + A A′ − AB

πA0 = π T +(yA0 − ye)

= πT +(AA′ − AB).

12Calculated using data on current price GDP ($US, billions) from the IMF World Economic Outlook database, October 2012.

11.4. APPENDIX 39

The central bank in bloc A, the central bank in bloc B and the foreign exchange market can now forecast that next period’s Phillips curve will be

πA1 = π A 0 +(y

A 1 − ye).

It’s also common knowledge that the central bank in bloc A has a monetary rule (MR) that defines the trade-off between output and inflation reductions each period – hence also next period, period 1:

yA1 − ye = −(π A 1 − π

T) (monetary rule; bloc A)

To simplify the notation, we assume α = β = 1. Putting period 1’s Philllips curve and monetary rule equations together, all three rational actors can work out the combination of inflation and output in period 1 in bloc A that the central bank of bloc A will want to see. Thus:

πA1 = π A 0 − (π

A 1 − π

πA1 = (πA0 + π

yA1 = ye − (πA0 − π

Now all three actors know the output level, yA1 , which the central bank in bloc A wants to achieve next period. To do so the only instrument bloc A’s central bank has at its disposal is rA0 . This has to be set to solve:

yA1 = A A′ − rA0 + q0

The problem is that q0 depends on r A 0 via the UIP condition:

rA0 − r B 0 = q

E 1 − q0

rA1 − r B 1 = q

E 2 − q

E 1

....

Summing both sides we get ∞�

(rAt − r B t ) = q̄

′ − q0 since lim t→∞

qEt = q̄ ′. So

we next need to model rBt . Since output in bloc B does not change (y B t = ye),

rBt , solves −(r B t − r̄

∗′)− (qt − q̄ ′) = 0 . Hence

∞�

(rAt − r B t ) = q̄

′ − q0

=⇒ ∞�

(rAt − r̄ ∗′)− (rBt − r̄

∗′) = q̄′ − q0

=⇒ ∞�

(rAt − r̄ ∗′)+

�

(qt − q̄ ′) = q̄′ − q0

40CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

Now assume that rA and q both converge to their new equilibrium levels r̄∗′, q̄′ at a proportional rate of λ. The derivation of λ follows the same method as in the small open economy (see Section 14.9.2 of Chapter 14).

L = (πA − πT)2 + β(yA − ye) 2

=⇒ (πA − πT)+ β(yA − ye) = 0

=⇒ (yA1 − ye) = − 1

β (πA1 − π

=⇒ (yA1 − ye) = − 1

1+ β (πA0 − π

From this we derive λ:

(yA1 − ye) = − 1

β (πA1 − π

T) and (yA1 − ye) = − 1

1+ β (πA0 − π

λ = (πA1 − π

(πA0 − π T) =

1+ β

So λ = 1 2 in the case that β = 1, it follows that:

rA0 − r̄ ∗′

1− λ +

q0 − q̄ ′

1− λ = q̄′ − q0

=⇒ rA0 − r̄

∗′

1− λ (q̄′ − q0)

� 1+

1+ λ

�

= (q̄′ − q0)

� 2− λ

1− λ

�

=⇒ rA0 − r̄ ∗′ = (q̄′ − q0)(2− λ).

This is now substituted into bloc A’s IS curve in deviation form to get:

(yA1 − ye) = −(r A 0 − r̄

∗′)−

� rA0 − r̄

∗′

2− λ

�

= −(rA0 − r̄ ∗′)

� 1+

2− λ

�

= −(rA0 − r̄ ∗′)

� 1− λ

2− λ

� .

This is the RX curve showing the relation along the equilibrium adjustment path – through the relevant points of IS curves with different values of q. Notably the RX curve is shallower than the representative IS curve, implying that a given change in r has a greater impact on y in the open economy than in the closed. This is because the change in r both operates directly on r with coefficient 1 (or more generally A), and operates indirectly on y via its effect on changing q with coefficient 1

2−λ . Thus a much smaller change in r is needed in

11.5. QUESTIONS 41

the open economy to have the same effect on y as in the closed economy. For example, if λ = 0.5, then r needs to change by only 1

3 of the amount as in the

closed economy. We should note also the ‘bloc’ effect. When q appreciates initially, rB has to

rise to keep yB in equilibrium. This requires a bigger change in rA than would be the case in the small open economy: in effect the rise in rA has triggered a rise in the world rate of interest, which would have been fixed in the small open economy. The empirical implication here is that we might expect to see common patterns to interest rate changes across the world if there is a shock in any one big bloc.

11.5 Questions

11.5.1 Checklist questions

1. Explain using words and diagrams how an oil shock can be considered both a demand shock and a supply shock.

2. Use Section 11.2.1 to answer the following questions about the oil price shocks of the 1970s:

a. How did the misdiagnosis of the oil shock of 1973/74 affect policy choices and economic performance?

b. Was the same mistake made following the 1979 oil shock? If possible provide some evidence to support your answer.

c. Use the AD −BT −ERU diagram to illustrate the basis of the policy error.

3. Oil prices fell dramatically in 1986. Use the W S−P S and ERU diagrams to explain the effect of this supply-side shock on a small open economy. At the initial real exchange rate, what has happened to real wages and the level of employment?

4. Assess the following statement: “The 2002-08 oil shock had less negative macroeconomic consequences than those in the 1970s due to the success of inflation-targeting central banks at stabilising their economies”.

5. Use an AD−BT −ERU diagram and the sector financial balances frame- work from Section 11.2.2 to show how a country whose government is restricting domestic demand could run ‘twin surpluses’ (i.e. government and current account surpluses).

6. A small open economy is initially in trade balance. There is a temporary increase in household income for one period (i.e. rise in �yt). What effect does this have on thecurrent account in period t andthe following periods? Does this story change if the increase in income is permanent?

42CHAPTER11. EXTENDINGTHEOPENECONOMYMODEL:OILSHOCKSANDIMBALAN

7. What is meant by the term ‘Dutch disease’? How does this relate to the intertemporal model of the current account?

8. Use Section 11.2.5 to answer the following questions about macroeconomic imbalances in the 2-bloc model:

a. Use a 2-bloc model to explain in words how there can be constant inflation in each bloc but current account imbalances.

b. Begin with 2 symmetric blocs. Now assume there is a permanent posi- tivedemandshockto blocAand anequal andoppositepermanentdemand shock to bloc B. Describe the new medium-run equilibrium (MRE). (Hint: Draw the AD − BT − ERU diagrams for the world, bloc A and bloc B before and after the shock (as in Fig. 11.13). Your answer should focus on the differences between the initial and new MRE. Don’t discuss the adjustment path to the new MRE). How could you adjust the nature of the shocks so that there was a lower real interest rate in the new MRE?

9. In the 2-bloc model, assume AA > AB. Show the mathematical derivation for the world real exchange rate and the world real interest rate.

10. Why is it not possible for all countries to follow a demand-focused (i.e. expansionary) growth strategy? How did differing growth strategies across economies with global impact help current account imbalances emerge in the pre-crisis period?

11.5.2 Problems and questions for discussion

1. Use the online series for real oil prices from the Carlin and Soskice web- page [insert address]. Pick an emerging and a developed economy and collect data from their national statistics or an international organization (e.g. IMF, OECD, Eurostat, World Bank) on the unemployment rate and the inflation rate from the start of the 1970s until the present. Do the patterns observed match those of the US in Fig. 11.9? If not, propose some potential reasons why?

2. Assess the following statement: "Economic policy makers should not di- rectly intervene to reduce current account imbalances, because they sim- ply reflect rational savings decisions and comparative advantages in an increasingly globalised world".

3. Set out the IS − RX and P C − MR diagrams for bloc A and bloc B (as per Fig. 11.15) and the associated impulse response functions (as per Fig. 11.16) for the case where there is a positive demand shock in bloc A and a negative demand shock in bloc B (assume the shocks are equal and opposite and take place simultaneously).

4. Optional: Use the mathematics from Section 11.4.1 of the Appendix to derive the RX curve after a negative demand shock in bloc B.