Degree Name

Doctor of Philosophy


School of Economics


Vanuatu's economic development has accorded monetary policy crucial roles related to price, monetary and financial-sector stability. Its role in price and monetary stability generally falls within the domain of monetary policy, and is determined by the ability of the Reserve Bank of Vanuatu (RBV) to control base money so as to create conditions conducive to economic development. It hinges on the notion that money plays a role in economic cycles.

The aim of the current study is to make an empirical examination for the first time, of monetary policy and its transmission mechanism in Vanuatu using three different econometric and dynamic models for the period that data is available, 1984-2009. Prior to the empirical study of the transmission mechanism, tests for unit-root and structural breaks using the endogenous Lagrange Multiplier (LM) method in Lee and Strazichic (2003, 2004) were carried out on the time-series data. Model C of the two-break test rejected the null hypotheses for a unit-root and structural breaks in nine of the 20 macroeconomic data series examined in this study. The tests found that structural breaks in real variables in Vanuatu occurred mostly after the 1990s, and tended to be associated with shocks in the international economy and with cyclones, while breaks in financial variables took place mostly in the latter part of the 1980s, and tended to be associated with domestic shocks.

The monetary transmission mechanism in Vanuatu was analysed using a small SVAR model with recursive and non-recursive identification restrictions. The study expanded the SVAR model to include disaggregated demand and sectoral GDP data to explore the outputcomposition channel of monetary policy. The second model applied in this study was a vector error-correction model (VECM), which uses Johansen's maximum-likelihood procedure to test for long-run cointegrating relationships. The final model used a simple new Keynesian dynamic stochastic general equilibrium (DSGE) model and introduced frictions in assessing monetary policy in Vanuatu for the first time. Compared to the backward-looking feedback rule of the SVAR model, the new Keynesian model incorporates forward-looking variables and solutions to account for expectations.

This study makes several contributions to the literature on the role of monetary policy and its transmission mechanism in Vanuatu. The findings show that structural breaks in real variables occurred mostly after the 1990s and tended to be associated with shocks in the international economy and cyclones, while breaks in financial variables took place mostly in the latter part of the 1980s and tended to be associated with domestic shocks. Similarly, shocks explaining movements in overall real output and disaggregated GDP components originate largely from external sector variables,in contrast to movements in domestic prices which are largely explained by variables in the domestic sector.

The traditional interest-rate channel to overall output and prices is weak. However, the interest rate through bank loans is relatively important at a more micro level, particularly with regards to aggregate-demand components such as gross investment, household consumption and other aggregate-demand components. In addition, foreign shocks, rather than domestic, play the more important role in transmitting shocks to domestic-deposit interest rates although they are not important in explaining movements in domestic-lending interest rates.

The credit channel of monetary policy through bank loans is found to be weak in transmitting shocks to disaggregated components of GDP but relatively important in transmitting shocks to nominal exchange rates. The study finds that the credit channel is driven by demand for loans rather than their supply.

The money channel to domestic prices is strong and persistent. The primary source of shock to prices is from reserve money. Furthermore, the money channel is important in transmitting shocks to nominal variables, notably bank credit and the nominal exchange rate. The findings support the use of reserve money, equivalent to the NBR in the US, as a variable representing shocks to monetary policy.

The money channel to overall output is weak, and is only relatively important to the specific disaggregated GDP and aggregate-demand components SERVSEC and HHC. In general, movements in overall GDP, aggregate-demand components and sectoral GDP components are explained by foreign shocks, including exchange rate shocks, rather than monetary shocks.

The exchange-rate channel is important in transmitting shocks to overall real output and to gross investment. The channel is strong, fast and persistent in transmitting shocks to financial and monetary variables, notably bank deposits, reserve money and broad money. By contrast, the exchange-rate channel to domestic prices is not strong.

The findings show important inter-sectoral linkages, particularly between household consumption and general government expenditure, and between the services sector and the industry sector. Shocks to the agriculture, fisheries and forestry sector of GDP have a large and fast impact on domestic prices. Price shocks constitute an important driver for changes in household consumption and the services sector of GDP. The study shows that following a monetary shock, the output-composition channel in Vanuatu works through investment rather than household consumption.

The study finds that an estimate of an appropriate weight between money and nominal interest rates as an instrument of monetary policy indicates a strong preference for money. This finding supports the current reserve-money targeting framework and use of reserve money, equivalent to the NBR in the US, as the instrument of monetary policy. In addition, the estimated mean monetary weight on inflation was found to range between 40% to 60% and 40% to 50% on output. Furthermore, frictions and other dynamics play important roles in macroeconomic fluctuations in Vanuatu.

FoR codes (2008)

140212 Macroeconomics (incl. Monetary and Fiscal Theory), 140303 Economic Models and Forecasting



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.