Monetary Policy and “Stabilisation Trap” in Selected Asian Countries – Implications for Poverty Reduction Anis Chowdhury Professor of Economics University.

Monetary Policy and “Stabilisation Trap” in Selected Asian Countries – Implications for Poverty Reduction

Anis ChowdhuryProfessor of Economics

University of Western Sydney, Australia

Introduction

… “deficit financing” can lead to a very unsustainable economy. Bolivia in the 1980s is an extreme example: its deficit rose to 28% of GNP – leading to hyperinflation and serious economic crisis. So, each country should aim at roughly balancing its budget…”

(Human Development Report, 1991, p. 42) • Generalisation from extreme cases led

to the policy package of IMF/World Bank in the 1980s & 1990s

Single digit Inflation and growth stagnation

Country  

Inflation Rate Growth Rate  

1965-80 1980-89 1965-80 1980-89  

Bangladesh 14.8 10.6 2.5 3.5  

Ethiopia 3.4 2.0 2.7 1.9  

Burundi 5.0 3.7 7.1 3.3  

Mali 9.0 3.6 4.2 3.8  

Niger 7.5 3.4 0.3 -1.6  

Rwanda 12.5 4.0 4.9 1.5  

Pakistan 10.3 6.7 5.2 6.4  

Indonesia 35.5 8.3 7.0 5.3  

Cote d’Ivorie 9.4 3.1 6.8 1.2  

Peru 8.1 5.6 4.1 2.1  

Congo 6.8 0.3 6.2 3.9  

Cameron 9.0 6.6 5.1 3.2  

Panama 5.4 2.5 5.5 0.5  

Gabon 12.8 -1.0 9.5 1.2  

Trinidad & Tobago 14.1 5.8 5.0 -5.5  

Poverty in the 1990s

Country $1 a day $2 a day

Bangladesh 29.1 77.8

Ethiopia 31.3 76.4

Burundi    

Mali 72.8 90.6

Niger 61.4 85.3

Rwanda 35.7 84.6

Pakistan 31.0 84.7

Indonesia 7.7 55.3

Cote d’Ivorie 12.3 49.4

Peru 15.5 41.4

Congo    

Cameron    

Panama 10.3 25.1

Gabon    

Trinidad & Tobago    

Argentina in the 1980s• hyper-inflation – the average annual inflation rate

around 391% and average annual economic growth rate around 1%

1990s• orthodox IMF stabilisation package succeeded;

average annual inflation rate 1.5%, and the economy grew by an average annual rate of 6.7% during 1991-1997

However…• the continuation of severe restrictive macroeconomic

policies forced the economy into a deflationary spiral – the inflation rate dropped to –0.9% and –1.3% in 2000 and 2001 respectively

• economy contracted by –3.4% in 1999 and by –2.1% in 2000

• unemployment rate rose from 6.5% in 1991 to 17.5% in 1996

• consequently, the poverty rate (head-count ratio) rose from 21.8% in 1993 to 34.3% in 2002, and the Gini coefficient from 0.45 to 0.49

Asian Experience• While lower inflation…is a positive indicator…a near-zero inflation rate

may be symptomatic of demand deficiency leading to capacity underutilization…Targeting for a too-low inflation rate…can sometimes result in overkill. .. Yet another problem with pushing inflation too low is that it will make it difficult to bring about the large relative price changes that the structural adjustment policies aim at. (Bangladesh Report, p. 38).

• The point is not that the authority has to fight inflation at any price and to only value deflationary policy. Rather, it has to face the tough question of how far to go with fighting inflation, knowing that with ongoing deflation an economy might face greater risk of entering into the chain of rising unemployment, falling demand, and reduction in the level of national income…This restricted policy should not be seen as the only way out, without flexibility. The inflation rate…has remained low since 1995… But, the targeted inflation of about 4 per cent…seems to be low if the aim is to create employment and economic growth. (Cambodia Report, pp. 48, 60-61).

• In the current context, one of China’s important challenges appears to lie in counteracting deflationary pressures and sustaining rapid economic growth rather than combating inflation...Persistent deflation may have serious adverse effects on China’s economic growth and poverty reduction prospects. (China Report, p. 72).

• Policymakers continue to adhere to tight IMF-prescribed fiscal and monetary targets in order to achieve single-digit inflation rates…Meanwhile, domestic consumption, not private investment, is supporting growth. But clearly this is not sustainable…High interest rates [needed for a low inflation target] are an impediment to growth in circumstances such as Indonesia’s, where the corporate sector is heavily indebted. (Indonesia Report, pp. 15, 19)

• Bank Indonesia highlighted Indonesia’s monetary policy dilemma: – “if the policy is tightly directed at attaining inflation target…it is feared

to disturb the current economic recovery.” – “If Bank Indonesia is still persistent in achieving the base money

target, extremely high increase in interest rate will be required, which…could harm the economic recovery prospects”

• This sentiment was echoed by the Chairman of the Indonesian Chamber of Commerce of Industry: “Bank Indonesia has been keeping interest rates high to ease pressure on the rupiah and stall inflation… however, [this] discouraged industries from making new investments that would otherwise have helped create new jobs”

Realisation at Last!

• Wolfensohn, “…if we take a closer look, we see something else – something alarming. In developing countries, excluding China, at least 100 million more people are living in poverty today than a decade ago. And the gap between rich and poor yawns wider.”

• World Bank (2005, p. xiii), “… there is no unique universal set of rules … [W]e need to get away from formulae and the search for elusive ‘best practices’…”.

• Easterly (2001) called the 1980s and 1990s “the lost decades”.

Monetary Policy and Poverty – theories and evidence

The channels through which monetary policy can affect poverty are:

• long-run economic growth • short-run output stabilisation and • income distribution.

Since monetary growth and inflation are positively linked, money’s role in poverty reduction will be examined by looking at the:

• inflation-growth inflation• short-run trade off between output and price

stabilisation and• inflation-inequality relationship. • We shall also examine the question of central

bank independence.

Money in the Long-run – Does Inflation Harm Economic Growth?

In theory, the answer is both Yes and No.

What is the empirical evidence?• Extensive empirical research indicates that the

negative relationship between inflation and economic or productivity growth is influenced by extreme values or outlying countries having an exceptionally high inflation rates (see Figure 2).

• In the words of Bruno and Easterly (1998), “the correlation loses significance with the omission of single observation – Nicaragua, which had hyper-inflation and negative growth in the 1980s … More generally, significance and sign of the cross-section correlation depends on the inclusion of the countries with high inflation crises – the above 40% episodes…”.

• They also observed, “the significance of the negative growth during 20-40% inflation vanishes if a single extreme annual observation is omitted – Iran in 1980”.

• Easterly (2003) reported similar findings and regarded inflation rates below 35% as “moderate”.

Figure 2a: Inflation-Growth (1980-1990)

-10

-5

0

5

10

15

-100 0 100 200 300 400 500

Inflation

Gro

wth

Figure 2c: Inflation-Growth (1980-1998)

-10

-5

0

5

10

15

-200 0 200 400 600 800

Inflation

Gro

wth

Figure 3: Average GDP Growth & Inflation Rates

(1960-2000)

-1

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

Between 0

& 5%

Between 5

& 10%

Between 10

& 15%

Over 15%

Average Inflation Rate

Ave

rage

GD

P G

row

th R

ate

(%)

1960-1979

1980-2000

Source: Islam, 2003

Figure 3 plots average inflation and growth rates over four decades (1960-2000) for 141 countries. Two interesting features emerge.

• The inflation-growth relationship was negative in the period 1960-1979. But it seems to have become non-linear – first positive and then negative in the later period of 1980-2000. This confirms the uncertain nature of the inflation-growth relationship.

• The variation in growth rate between the two time periods is much larger at both low and high inflation. That is, both high and low inflation rates can be destabilising, and hence harmful for the poor.

• Thus, one may conclude that while policy makers should guard against high inflation, a country may need “moderate” inflation to sustain economic growth. This understanding is essential when there is excess capacity and persistent high unemployment or underemployment.

• The over-fixation with a single digit inflation target cannot be justified based on the fear of inflation going out of control once it is allowed to go beyond, say 10% level.

• Dornbusch and Fischer (1993) found that inflation rate in the moderate range of 15-30% does not usually accelerate to extreme levels.

• Bruno and Easterly (1998) found that the threshold inflation rate of 40% at which the probability of inflation rate accelerating rises significantly.

• Furthermore, only in a handful of cases inflation rate did accelerate and output stagnated or declined in the past, and these cases could be attributed to unusual circumstances (e.g Iran or Nicaragua in the 1980s following dramatic fall of the regimes).

• Cross-country scatter diagram (Figure 5) shows, the claim of the orthodox school that inflation harms the poor breaks down when one considers inflation rate in the range of 5-20%. Consistent with the aggregate inflation-growth relationship, the negative relationship between inflation and the income of the poor is based on few cases of extreme inflation.

• Therefore, contrary to the orthodox view, poverty-reduction strategy requires moderately expansionary monetary policy, and “fear” of excesses cannot be a basis for sound public policy.

Money in the Short-run: Is there any Output Stabilisation Role?

Demand Shock: Monetary policy’s role in stabilizing output depends on two questions:

• To what extent a country is prone to demand shocks

• To what extent prices (product prices, exchange rate & interest rate) and wages flexible

The components of aggregate demand are less stable in developing countries due to:

• Narrow export base• Poverty and the role of current income• Liquidity constraint for both households & small

firms

Among key prices• Exchange rate is quasi fixed• Wages are sluggish downwards – mainly

because real wage is already too low.

Therefore, regardless of whether the adjustment happens through output or prices, falling aggregate demand will have serious implications for the poor.

• If the adjustment happens through cuts in output and employment, the first to lose job is unskilled and unorganised labour.

• If the adjustment happens through declines in wages, again the unorganised and unskilled workers would be forced to accept lower wages.

S

Po

P1Do

D1

Y2 Y1 Yo

Figure 6: Adjustments to Demand Shocks

Thus, the poor unorganised workers are forced to choose between jobs and lower real wages. In either case, the average income of the poor is likely to drop when nominal GDP growth drops from its trend. Cross- country (Figure 7) evidence shows that average income of the poor is negatively related to aggregate demand variability.

• This negative relationship does not breakdown even when the outliers are omitted.

• Therefore, from the point of view of protecting the poor, monetary policy needs to stabilise output in the face of adverse demand shocks.

Supply Shock: It is generally accepted that the developing countries are more prone to supply shocks than demand shocks.

• This is due to their heavy dependence on agriculture and imported raw materials, and energy (oil).

• While alternating floods and drought are almost regular phenomena affecting agriculture, declining terms of trade due to rising prices of imported raw materials and energy adversely affects the industrial sector.

• The choice between output and price stabilisation becomes starker in the case of a supply shock. This can be shown diagrammatically:

Figure 8: Adjustment to Supply Shocks

D1 S1 S1Do So Do So

D1P2

P1 P1Po Po

Y1 Yo Y2 Y1 Yo

• In panel A, the response to an adverse supply shock is an expansionary monetary policy to stabilise output at Q0.

• In panel B, the response is a contractionary monetary policy to stabilise the price level at P0.

• When the response is an expansionary policy, the price level rises further to P2, causing higher inflation.

• When the objective is price stabilisation with a contractionary monetary policy, output declines further to Q2.

Inflation and Income Distribution

Inflation can disproportionately hurt the poor through two channels:

• Wage rise lags behind price rises• Poor mostly save in money and inflation reduces

the value of their savings.

Counter arguments:• If real wage declines due to inflation then

employment should rise. Therefore, the employment effect of inflation can outweigh the real wage effect on poverty.

• The gain from expansionary monetary policy will not be temporary if the inflation rate remains “moderate”. This is evident from the positive relationship between moderate inflation and economic growth as demonstrated earlier.

• The poor are largely net financial debtors. Thus, inflation can benefit the poor by reducing the real value of their net debt. On the other hand, lower inflation not only increases the real value of financial debt, the high interest rate policy aimed at bringing inflation down increases the debt servicing cost of indebted poor. This makes them, doubly disadvantaged.

Evidence

• Studies have found that income distribution narrows during the expansionary phase and widens during the contractionary phase of a business cycle.

• Figure 9 presents the scatter plots of average inflation and inequality (measured by Gini coefficient). Once again, we find that the claim of adverse distributional effect of inflation is based on extreme inflationary cases.

• There seems to be no relation between inequality and inflation when the inflation rate ranges between 5 and 15%. On the other hand, inequality rises with the variability of nominal GDP growth (Figure 10).

• The evidence presented in Figures 9 & 10 vindicates the need for output stabilising monetary policy that allows for “moderate” inflation.

Central Bank Independence and Inflation-targeting

• The empirical evidence on the performance of independent central banks is still mixed.

• The conditions required for the success of an inflation-targeting approach include– the lack of fiscal dominance and – the absence of any other objectives.

• None of these conditions appears to hold in most developing countries. – The revenue base of these countries is very low

and their capital market is underdeveloped. This forces most developing countries to borrow from the central bank.

– These countries also have some sort of quasi-fixed exchange rate systems needed to prevent imported inflation and to attract short-term portfolio foreign capital.

– Thus money supply responds to developments in government finance and the balance of payments.

Leaving aside the technical argument, there is a broader issue of democratic governance and technocratic insulation of institutions.

• It is pertinent at this juncture to quote Milton Friedman who is on record for voicing the concern '… money is too important to be left to the central bankers'. His concerns are elaborated in the following quote:

“The political objections are perhaps more obvious than the economic ones. Is it really tolerable in a democracy to have so much power concentrated in a body free from any direct political control? … One economic defect of an independent central bank … is that it almost invariably involves dispersal of responsibility… Another defect … is the extent to which policy is … made highly dependent on personalities… A third technical defect is that an independent central bank will almost invariably give undue emphasis to the point of view of bankers… The defects I have outlined constitute a strong technical argument against an independent central bank.”

Stern and Stiglitz (1996) have made the point more succinctly:

“The degree of independence of the

central bank is an issue of the balance of power in a democratic society. The variables controlled by the central bank are of great importance and thus require democratic accountability. At the same time the central bank can act as a check on government irresponsibility. The most successful economies have developed institutional arrangements that afford the central bank considerable autonomy; but in which there is a check provided by public oversight, an oversight that ensures the broader national interest is taken into account in the final decisions.”

In this respect, it is worth highlighting the distinction between "goal independence" and "instrument independence".

• The former refers to central bank's ability to set the inflation target independently of the government.

• The later is its independence in the choice of instruments and hence relates to central bank's day to operations.

• No central bank can be entirely independent of a democratic government while it can be entirely free in choosing its instruments.

• Most developing countries are new democracies. In such a situation, a central bank with both goal and instrument independence may choose a very low inflation target which can undermine a nascent democracy by delaying economic recovery.

• Furthermore, the very argument that an elected government cannot be trusted with the responsibility of managing the economy goes against the very principle of representative democracy.

The above issues have become more prominent in Indonesia as the new central bank law (enacted in 1999) grants the Bank Indonesia (BI) both goal and instrument independence.

• This has been responsible for open disputes between the BI and parliament in a number of occasions on the appropriateness of monetary policy stance.

• The national planning agency (BAPPENAS) has also expressed concerns about the mismatch between the monetary policy stance of the BI and fiscal policy.

The Indonesia country report (p. 19) observes:• It is difficult to ease monetary policy and achieve

some consistency between fiscal and monetary policies when the Bank of Indonesia (BI) remains relatively autonomous – and wedded to tight monetary policies. While BI should have autonomy in determining its policy instruments, its objectives should be subject to public discussion and oversight. By setting low inflation as its overriding objective, BI can compromise the achievement of other objectives, such as growth of income and employment generation, which most people value highly, in addition to price stability.

It is rather strange that under the provision of the current legislation of BI independence, neither the president nor the parliament can remove the governor of BI before the expiry of his/her tenure.

• This led to the much publicised stand-off when President Wahid wanted to remove the Governor after he was indicted in a corruption case.

• The Governor refused to resign even when he was convicted following a guilty verdict.

In sum, inflation targeting and central bank independence are not merely technical matters, as the orthodoxy tends to believe.

• It is pertinent at this juncture to point to the observation made by a central bank insider, Guy Debelle (1996: 1).

“An increase in the inflation aversion of

the central bank, while always reducing inflation rate, may reduce welfare because of its adverse effects on output and government spending. The net welfare effect is shown to depend on the weights in the welfare functions of the fiscal authority and society. Thus, increasing the central bank's inflation aversion is not necessarily a free lunch.”

Thus, the essence of inflation targeting is embedded in the so-called social welfare function that includes both inflation and economic growth.

• High unemployment that is required to bring inflation rate to a single digit level or to keep inflation rate in the range of 3-5% has significant and systematically regressive effects on the distribution of income.

• The poor fare worse when unemployment rises and persists, especially when there is no adequate safety-net or social security system. At the same time the real value of their net debt rises with falling inflation.

• Hence the poor have reasons to be more averse to unemployment and less averse to inflation than the elite in society.

• As the poor lack voice and representation, the choice of weights for inflation and unemployment in the social welfare function raises an important issue of conflicts and political economy of public policy.

Policy Recommendations

• Recognise both price and output stabilisation roles. That is, avoid both too conservative and too expansionary monetary policy. Inflation in the range of 10-20% can be regarded as safe both from the point of view of avoiding a stabilisation trap and harmful affects of expansionary monetary policy.

• Achieve consistency with the fiscal policy stance. Safe expansionary monetary policy within the above guideline will allow governments to borrow from the central bank to finance employment-intensive public investment programs in infrastructure. This can create and stabilise employment without pressure on interest rate. Thus private investment is unlikely to be crowded out; rather both domestic and foreign investment will be encouraged by demand growth and externality benefits from improved physical infrastructure.

• Develop directed credit programs to employment-intensive small and medium enterprises, agriculture and rural industries. This is essential because they are more dependent on bank credits than larger enterprises that have better access to capital markets. Therefore, even when over-all credit growth needs to be restrained, directed credit to SMEs and rural-agricultural sectors must be maintained to avoid asymmetric adverse impact on employment. This will protect the income of the poor and offset likely adverse impacts of cyclical downturn on inequality.

• Central banks should be given autonomy to choose and implement the instruments of monetary policy within the over-all economic objectives dictated by the poverty reduction strategy of the government. This means a participatory policy making process so that the trade-off parameter between inflation and unemployment reflects the concerns of the poor, and not of the elite or multilateral agencies.

Given that there is no evidence of a trade-off between employment creation and moderate inflation, the conflict between goals and instruments is not as stark.

Yet for two targets the monetary authorities can have two instruments:

• Traditional instrument of interest rate (or such instruments as reserve requirements) assigned to keep inflation at a moderate level.

• Specialised credit regulation directed to employment creation.

The central banks can consider a number of options in designing specialised credit programs:

• Follow the Indian example, where all banks (public and private) are required to lend at least 40 per cent of their net credit to the “priority sector”. If banks fail to meet this requirement, they are required to lend money to specific government agencies at a very low interest as a penalty.

• Alternatively, use some carrot and stick measures by combining the penalty system with incentive based measures such as asset based reserve requirements, support for pooling and underwriting small loans, utilising the discount window in support of employment generating investments

Asset based reserve requirements are an effective tool for creating incentives for banks to invest in socially productive

assets. For example:

• Based on well-research findings of employment elasticities, the central banks would list a set of employment generating investment, and a lower reserve requirement would apply for the deposits invested in these activities than the deposits invested in speculation or Treasury Bills.

• The central banks can also take steps to create liquidity and risk sharing institutions for loans to small businesses which have promise to generate employment, but do not have adequate access to the credit market. For example, the central banks can provide financial and administrative support for asset backed securities which would take loans to small businesses and other employment intensive activities, bundle these investments and sell them as securities on the open market.

• Finally, the central banks can open a special discount window facility to offer credit, guarantee or discount facilities to institutions that are on-lending to firms and co-operatives engaged in employment intensive activities.

To achieve the RER target, the central banks should intervene in the exchange market.

That is; the nominal exchange rate should move to hold RER at a stable and competitive level for an extended period of time.

There are basically three options: 1.      interest rate manipulation

2.      sterilised intervention3.      capital controls.

To support the developmental role of exchange rate, monetary policy must maintain stable and low real interest rates. By boosting exports this will complement the employment target of monetary policy.

Will low interest rates set off inflationary nominal depreciation (under speculative exchange rate dynamics)?

 Targeting RER can help central banks

avoid this problem.

Finally central banks need to have some controls on capital flows. This will give central banks controls over monetary aggregates and hence monetary policy independence to keep real interest rates low, stabilise employment and keep inflation at a moderate level, while the exchange rate policy aims to maintain international competitiveness.

• There are, of course, many critics of capital controls. However, they must accept that the Mundell-Fleming model conceived of capital flows as largely money-market flows or at most money and bond markets flows.

• An important development in the world economy in the late 1990s was the shift of international capital flows from the fixed income market – both money and bond flows – to the equity market – both portfolio equity flows and FDI.

• A decline in policy interest rates can raise expected corporate earnings. This can lead equity prices to rise and attract foreign investors with extrapolative expectations to buy more equities. Therefore, equity effect of lower interest rates can be larger than the money-bond market effects to overturn standard Mundell-Flemming results.

• Thus, capital account openness should not be viewed as an all-or-nothing proposition. The increased importance of equity flows has increased the effective scope of a capital account policy of semi-openness. A capital account can be open to equity flows – both portfolio and FDI, but closed to money and bond flows.