Inflation targeting in developing countries revisited

Posted by Anis Chowdhury on 17 March 2011

Inflation targeting in developing countries revisited


Sarah Anwar, Anis Chowdhury and Iyanatul Islam[1]


Inflation targeting (IT) has been the dominant monetary policy paradigm since 1990.[2] There are now 17 emerging and developing economies that practice IT with a median targeted inflation rate of 3%.[3] A recent review observes[4], “A growing number of countries are making a specific inflation rate the primary goal of monetary policy, with success”. This ‘success’ is summed up as follows:

  • Both inflation-targeting and non-inflation-targeting low-income economies experienced major reductions in inflation rates and improvements in average growth rates. Although the non-inflation-targeting countries continued to have lower inflation and higher growth than the inflation targeting countries, those that adopted inflation targeting saw larger improvements in performance.
  • Both inflation-targeting and non-inflation-targeting low-income economies also experienced large reductions in the volatility of inflation and output, with the countries that adopted inflation targeting registering bigger declines, especially in inflation volatility.


One may note the following about above findings. First, in terms of prevailing inflation and growth rates, non-IT countries do better than IT countries. Second, the larger improvement in IT countries is possibly due to the fact they  had higher median initial inflation (16% vis-à-vis 10% in NIT) and lower median initial growth rate (3% vis-à-vis 4.5% in NIT). Inflation declined in both IT and NIT countries, implying that central banks perhaps do not have to have an explicit quantitative target for inflation. Furthermore, there is strong evidence that the decline in inflation might not be due to IT itself, but rather to the general decline in world-wide inflation or to a simple reversion to a more normal inflation rate (Ball and Sheridan, 2003).


The case for revisiting inflation targeting is timely. Distinguished participants at a recently held conference (7-8 March, 2011) at the IMF argued the case for a “wholesale re-examination of macroeconomic policy principles” in the wake of the Great Recession of 2008-2009. In particular, both the Managing Director of the IMF and the Director of the IMF’s Research Department have noted that the IT paradigm was dominant in the pre-crisis period and that it needs to be revisited.[5]


Inflation-growth relationship


The nature of inflation-growth relationship is not straight forward.[6] As Milton Friedman (1973: 41) reminded us, “Historically, all possible combinations have occurred: inflation with and without [economic] development, no inflation with and without [economic] development”. In a seminal paper, Bruno and Easterly (1998: 3) asked “Is inflation harmful to growth?” and based on their cross-country econometric analysis concluded, “The ratio of fervent beliefs to tangible evidence seems unusually high on this topic.”


Using data from 140 countries (comprising both developed and developing countries) from 1960 to 1998, Khan and Abdelhak (2001) found that the threshold level of inflation above which inflation significantly slows growth is estimated at 11 to 12% for developing countries.[7] Based on non-linear regression estimates of the relationship between inflation and economic growth for 80 countries over the period 1961-2000, Pollin and Zhu (2006) found that higher inflation is actually associated with moderate gains in GDP growth up to a threshold of approximately 15 to 18% inflation. The study by Sepehri and Moshiri (2004) of panel data from both developed and developing countries show that the estimated turning points varied widely from as high as 15% per year for the lower-middle-income countries to 11% for the low-income countries, and 5% for the upper-middle-income countries. 


A 2010 study by Brito and Bystedt (2010) used some advanced econometric tools to correct some major limitations of the previous analyses and found that IT actually resulted in lower output growth during adoption.[8] The reduction in output growth means that IT policies do seem to hinder economic growth, at least in the period studied. Moreover, if economic growth is slowed, then the rate of growth of potential output is also lowered.[9]


Other ways in which IT may negatively affect the trajectory of growth and development include high interest rates, real exchange rate and financial volatility. In many emerging and developing countries, maintaining very low inflation rates requires significant increases in the real interest rate (Epstein, 2008). High real interest rates affect investment and may have negative consequences for growth and development. Moreover, in economies with relatively unrestricted capital mobility and reasonably developed capital markets, the high interest rates associated with IT often attract inflows of short-term portfolio investment. Such capital flows can lead to an appreciation of the real exchange rate, hurting exports and facilitating import penetration (Galindo and Ros, 2008). Tradable sectors will be negatively affected by the appreciation, leading to a reallocation of resources to the non-tradable sector. If productivity levels, on average, are lower in the non-tradable sector, the outcome will be slower growth and delayed industrialization. Additionally, the accumulation of stocks of short-term capital increases the risk of financial fragility. A rapid reversal of these flows can lead to a collapse of the currency and, in turn, a broader economic crisis, thereby placing the IT country in a fragile position.


Sources of inflation matter


Most developing countries are prone to supply shocks due to their high dependence on agriculture and imported energy. A classic case is the food and energy price shocks that badly hit developing countries in the late 2000s. Today, high and rising food prices pose a major policy challenge. Indeed, the correlation coefficient between median inflation rates in LDCs (least developed countries) and a global food price index is 0.82.[10] One estimate suggests that about 44 million people might be pushed into at least transient episode of poverty as a result of high and rising food prices.[11]


Supply-side shocks may simultaneously reduce growth and raise inflation. Tightening monetary policy in response to this kind of shock may make the situation worse (Friedman and Kuttner, 1996; Chowdhury 2005). Output fluctuations will be greater when macroeconomic policies remain focused on price stability in the face of such shocks as the burden of adjustment falls on only one variable (output). That is, strict IT might introduce a pro-cyclical bias into monetary policy for countries in which supply-side inflation is commonplace. The degree of this bias will depend on the relative importance of supply-side factors in determining inflation and the amount of discretion monetary authorities are allowed.


The use of monetary tightening in response to supply shock inflation produces greater output growth volatility.  There is a growing body of empirical research that finds a robust, negative cross-country relationship between growth and growth volatility. They also find a significant negative correlation between growth and medium-term business cycle fluctuations. (See for example, Ramey and Ramey, 1995; Kroft and Lloyd-Ellis, 2002).


Therefore, IT must be flexible enough to respond differently, depending on the source of inflation. The role that monetary policy can play in dealing with supply shocks is strictly limited and central banks should refrain from using the policy interest rate to deal with such supply side forces, especially when the inflation surges are accompanied by food price increases. Interventions by the government to enhance food security represent much more appropriate responses.


IT beyond growth


When the superiority of IT in terms of growth performance cannot be ascertained unequivocally, this section focuses on other indicators, such as employment and poverty. We compare 12 IT countries with 12 non-IT countries with similar characteristics, such as Human Development Index scores, level of income per capita and being in the same or a nearby region.[12]


While a number of factors, especially labour market institutions, can affect labour productivity and other labour market indicators, including poverty, we find interesting association between IT or NIT and the median values of these indicators for the period 2000-2007. For example, median labour productivity is higher in NIT than in comparable IT countries – $14,999 in NIT vs. $14,027 in IT in constant 1990 PPP$. While there is not much difference in the median unemployment rate (around 10%) and median poverty rate (around 27%) between IT and NIT countries, vulnerable unemployment is higher in IT countries – around 37% – than in NIT – around 30%. Higher vulnerability may be due to pro-cyclicality of the IT regime. This also shows that in the IT regime, the burden of adjustment might fall mainly on labour.


Concluding remarks


There is ample evidence to suggest that targeting low, single-digit inflation is not necessarily a good development strategy.  One should distinguish between the need to safeguard price stability as a principle and the more restrictive notion of targeting a specific inflation rate. One should go back to the refreshing eclecticism of the founding fathers of the IMF. As the preamble of the IMF’s Article of Agreement IV notes: “… each member shall (i) endeavor to direct its economic and financial policies toward the objective of fostering orderly economic growth with reasonable price stability, with due regard to its circumstances”. The preamble not only expects monetary policy to attain simultaneously both reasonable price target and orderly growth, but also, contrary to the IT regime, it does not specify any specific quantitative target. Additionally, there is no presumption of the suitability of one target (less than 5%) that is universally applicable as due regard needs to be given to country specific circumstances.




Ball, Laurence and Niamh Sheridan (2005). “Does inflation targeting matter?” in B.S. Bernanke and M. Woodford, eds. The Inflation Targeting Debate. Chicago: University of Chicago Press, pp. 249-76.


Banerjee, Abhijit and Esther Duflo (2008). “Do Firms Want to Borrow More? Testing Creit

Constraints Using a Directed Lending Program,” Manuscript, Massachusetts Institute of



Brito, Ricardo and Bystedt, Brianne (2010). “Inflation targeting and emerging economies: panel evidence”. Journal of Development Economics 91: 198-210


Bruno, Michael and William Easterly (1998). “Inflation crises and Long-run Growth”. Journal of Monetary Economics, 41: 3-26.


Chowdhury, Anis (2005). Thematic Summary Report: Monetary Policy. UNDP Asia-Pacific Regional Programme on the Macroeconomics of Poverty Reduction.


Epstein, Gerald (2008). “An employment targeting framework for central bank policy in South Africa”. International Review of Applied Economics, 22(2): 243-58.


Friedman, Benjamin and Kenneth Kuttner (1996). “A Price Target for U.S. Monetary Policy? Lessons from the Experience with Money Growth Targets”. Brookings Papers on Economic Activity, 27(1): 77-146.


Friedman, Milton (1973). Money and Economic Development, Toronto: Lexington Books


Galindo, Luis Miguel and Ros, Jaime (2008). “Alternatives to inflation targeting in Mexico”. International Review of Applied Economics 22(2): 201-14.


Khan, Mohsin and Senhadji Abdelhak (2001). “Threshold Effects in the Relationship between Inflation and Growth”. IMF Staff Papers, 48(1)


Kroft, Kori and Huw Lloyd-Ellis (2002). “Further Cross-Country Evidence on the Link between Growth, Volatility and Business Cycles”. Queens University Working Paper.


Pollin, Robert and Andong Zhu (2006). “Inflation and Economic Growth: A Cross- Country Non-linear Analysis”. Journal of Post Keynesian Economics 4: 593-614.


Ramey, Garey and Valerie Ramey (1995). “Cross-Country Evidence on the Link between Volatility and Growth”. American Economic Review, 85(5): 1138-51


Sepehri, Ardeshir and Saeed Moshiri (2004). “Inflation-Growth Profiles Across Countries: Evidence from Developing and Developed Countries”. International Review of Applied Economics, 18: 191–207.


Roger, Scott (2010). “Inflation Targeting Turns 20”. Finance & Development, March.


[1] Sarah Anwar and Iyanatul Islam, ILO, Geneva; Anis Chowdhury, UN-DESA, New York and University of Western Sydney, Australia. The views expressed here are strictly personal and do not necessarily reflect the views of the United Nations or any of its agencies/funds/programs.

[2] New Zealand was the first country to formally adopt an inflation target of 0-3% in March 1990.

[3] Authors calculation based on available data from central bank websites.

[4] Roger (2010).

[5] ‘IMF triggers debate on crisis lessons’, IMF Survey online, March 8, 2011 available at

[6] Proponents of IT sometimes hold seemingly contradictory views. One the one hand they claim that inflation is largely a monetary phenomenon and thus independent of the growth rate which is driven by ‘real’ forces in the long run. On the other hand, they suggest that rising inflation rates damage growth and employment prospects.

[7] They also note that “The positive effect of inflation on growth is only present for inflation rates lower than…. 18 % for developing countries.” (p. 16). This implies that the upper bound is 18 %.

[8] The Brito and Bystedt study focused on a panel sample of 46 developing countries (13 IT countries) between 1980 and 2006.

[9] That is, potential output depends on actual output. This is what one can expect from the “learning by doing” effect within the endogenous growth model. When an economy operates at its potential, it creates more potential as working people become more skilled through learning on the job. The Keynesian acceleration effect also operates as business invests in capacity building as people’s purchasing power rise with the rise in their income. On the other hand, when the policy framework keeps the economy under its capacity, it lowers the potential as investment remains depressed and people become deskilled through long-term unemployment.

[10] Authors estimates

[11] World Bank latest issue of ‘Food Price Watch’ available at price watch report, February 2011.

[12] The developing IT countries are : Brazil, Chile, Colombia, Guatemala, Mexico, Peru, Philippines, Indonesia, Ghana, Thailand, Turkey, South Africa. The comparable non-IT countries are : Argentina, Ecuador, Honduras, Uruguay, Panama, India, Jordan, Kenya, Sri Lanka, Lebanon, Botswana. The choice of this sample is no more or no less arbitrary than other selection procedures.