Essays on monetary policy with sterilised intervention in emerging market economies

Devadas, Sharmila (2013) Essays on monetary policy with sterilised intervention in emerging market economies. PhD thesis, University of Nottingham.

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB) | Preview


The purpose of this thesis is to provide a new simple theoretical framework for understanding the sterilisation of foreign exchange intervention within a monetary policy framework where a short-term interest rate is the main policy instrument. Another aim is to carry out new empirical analysis that compares and contrasts the effects of intervention on base money growth with its effects on broad money growth, and to investigate if these effects are associated with common country characteristics in terms of the nature of intervention, balance of payments flows and monetary policy frameworks.

Using a portfolio balance model and simple balance sheet constraints, it is shown that base money sterilisation does not imply broad money sterilisation. Incomplete broad money sterilisation, when intervention is a positive money supply shock, leads to looser monetary and financial conditions, even as base money growth is completely sterilised in an interest rate-targeting framework. With a progressively lower degree of broad money sterilisation, the policy interest rate in an optimal monetary policy reaction function needs to respond more strongly to the exogenous factors affecting the output gap and inflation.

The empirical analyses of the contemporaneous and long-run effects of real intervention on real base money growth and real broad money growth are carried out using multivariate autoregressive distributed lag (ARDL) regressions for individual countries. These regressions include controls for demand factors. The focus on a reasonably large sample of 30 countries, with individual country heterogeneity taken into account, and the emphasis on broad money sterilisation represent a clear departure from existing empirical literature.

The empirical results reveal that on average, the real intervention effect on real broad money growth is higher than and disconnected from the effect on real base money growth. The baseline group average short-run and long-run coefficients are 0.122 and 0.496 respectively for real broad money growth and 0.075 and 0.111 respectively for real base money growth (excluding outliers, Egypt and Taiwan, in both instances). With regard to the relevance of country characteristics, a general lack of statistical significance is observed across mean and median equality tests and bivariate regressions. This is not unexpected with regard to real intervention effects on real base money growth, but indicates that the real intervention effects are very much country-specific in the case of real broad money growth. In particular, there is no evidence of a difference in real intervention effects on real broad money growth between inflation-targeting and non-inflation-targeting countries. The short-run and long-run real intervention effects on real base money growth are found to be robustly associated with differences in capital account openness although these differences are not of a monotonic nature. Meanwhile, the short-run real intervention effects on real broad money growth have a positive, monotonic association with current account surpluses and concurrently a negative monotonic association with capital account surpluses. On the other hand, the long-run real intervention effects on real broad money growth do not appear to be robustly linked to any particular country characteristic in a statistically significant manner.

A key implication of the thesis is that broad money sterilisation matters, not base money sterilisation, in understanding how balance of payments flows and intervention permeate the economy. Complete broad money sterilisation, however rarely occurs. Furthermore, regardless of the degree, broad money sterilisation, seen as a blunt instrument, may have unpredictable and undesirable effects, owing to uncertainties over money demand and money supply shocks, and mismatches in asset demand and supply. The model presented provides a framework for understanding the continued reliance on measures such as prudential policies and restrictions on capital flows in dealing with financial imbalances; this despite seemingly successful sterilisation by way of the containment of base money growth.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Bleaney, M.F.
Roberts, M.A.
Keywords: Monetary policy, foreign exchange rates
Subjects: H Social sciences > HG Finance
Faculties/Schools: UK Campuses > Faculty of Social Sciences, Law and Education > School of Economics
Item ID: 13087
Depositing User: EP, Services
Date Deposited: 02 Oct 2013 13:53
Last Modified: 14 Sep 2016 10:20

Actions (Archive Staff Only)

Edit View Edit View