Degree Name

Doctor Of Philosophy


School Of Economics and Information System - Faculty of Commerce


The major aims of this study are as follows. First, to macro model prospective developments in the Lebanese economy for policy analysis and evaluation. The main purpose of this study is to develop a dynamic macroeconomic model for Lebanon including the budget deficit and the funding of it, as well as the composition of government expenditures (capital or current). Hence this study develops behavioural equations not used before for Lebanon. The model developed emphasises the effects of exogenous shocks arising from budget deficits and the funding of it (e.g. monetary accommodation or bond financing). This macroeconomic model is utilised as well to analyse the effects of exogenous shocks arising from increased government expenditures (capital expenditure or consumption expenditure) upon key macroeconomic variables. The current government’s policy approach as well as the separate government policies in response to the Lebanese fiscal crisis is analysed as well through the use of this macroeconomic model. The second aim of this study is the application of estimation techniques to Lebanese data. This study generates parameter values of key macroeconomic behavioural relationships in the context of Lebanon by using efficient estimation techniques and appropriate data definitions for Lebanon. This represents a new contribution to the literature, since there have not been any empirical studies on such relationships in the case of Lebanon. The third aim of this study is the application of a simulation analysis to the Lebanese economy. This study conducts a numerical simulation analysis of the macroeconomic model developed, in order to analyse a number of economic policies in the context of the Lebanese fiscal crisis with the aim of improving the country’s macroeconomic performance. Since there has been no research in the area of simulation and policy analysis for the case of Lebanon, this research makes a major contribution to the literature and to an understanding of the Lebanese economy. The model developed in this study is based on the contributions of the Dornbusch model (DB) (1976), the portfolio balance model (PBM) (Branson (1977, 1984) including the work of Dornbusch and Fisher (1980). The model also ncorporates the work of Harvie and Kearney (HKM) (1996). The DB and PBM models have a number of deficiencies, especially the neglect of the supply side of the economy; no wealth effects; no funding implications from fiscal deficits; and a lack of stock-flow interactions, among others. The model developed attempted to remedy these deficiencies by incorporating the supply side of the economy, wealth effects, capital stock accumulation, alternative ways of funding the budget deficit and the composition of government expenditures (capital or current). Hence, many amendments have made to these existing models in order to make the model developed more applicable to the case of Lebanon, especially to analyse the impact of the composition of budget funding and the composition of government expenditures shocks on macroeconomic variables. Because of a lack, or unavailability, of parameter values for key macroeconomic behavioural relationships in the context of Lebanon, and the need for these parameters values in the conduct of a simulation analysis, this study empirically estimated the behavioural equations of the macroeconomic model developed by using efficient testimation techniques (FIML, ECM) and appropriate data definitions for Lebanon. The results from the FIML approach indicated the existence of long run relationships, or cointegration, for all the equations employed. The findings of cointegration allow us to investigate the dynamics of the system with the information of the cointegration relationship, and an error-correction model (ECM) is estimated. Hence these estimates (the long run and short run estimated coefficients from cointegration and the error correction model) provide a range of possible parameters values, which are used to conduct a simulation and policy analysis in chapter 7. The parameter values estimated make a unique contribution to the literature, since there are no other empirical studies analysing such behavioural relationships in the case of Lebanon. In addition, this study examined the time series data to find out whether the series are stationary or non-stationary, by using the Augmented Dickey-Fuller (ADF) test. The results showed that only real private consumption expenditure ( c p ), real government investment spending ( i g ), rate of change of real bonds stock (b&− p&) , real government expenditure (g) and real private investment spending ( i p ) seem to be stationary I(0). The rest of the variables included in the model are found to be non-stationary of different orders, either I(1) or I(2). The major findings from the simulation results presented in this study are that, implementing the policy of expansion in government capital expenditure, for two presumed cases (unanticipated/gradual), produces larger favourable impacts upon Lebanese economic development in terms of private sector investment, and in terms of the supply side of the economy (crowding in effects) during the whole adjustment process towards long run steady state. This policy produces, as well, favourable impacts in terms of external developments. Implementing the policy of an expansion in government consumption expenditure produces unfavourable effects in terms of external developments during the djustment process. The trade balance deteriorates in line with a deterioration in foreign asset stocks as a result of current account deficits, and hence results in an increase in foreign debt. This policy produces, as well, unfavourable effect in terms of private investment and aggregate supply (crowding out effect). Implementing the two policies (expansion in capital expenditure/government consumption expenditure) produces similar outcomes in terms of the interest rate and the rate of inflation. However, both policies produce higher inflation during the short run period. The interest rate is higher as well during the first year of the short run period due to the increase in public spending arising from the funding component through bond sales. However, the simulation results for the two policies show that money deficit financing is inflationary and shows large sensitivity in terms of the interest rate. Bond financing is non inflationary and shows little sensitivity in terms of interest rates. The main finding is that if the government considers a fiscal expansion policy in order to improve macroeconomic performance, the simulation results suggest that the government should adopt the policy of an expansion in capital expenditure because it produces the most desirable outcomes. In addition, it should adopt a gradual approach because this produces considerably less volatility in terms of major macro variables. The main findings from our simulation results dealing with the government approach to the fiscal crisis, does not support the government policy in dealing with the crisis. The results presented suggest that it produces the most undesirable economic outcomes, and hence will only exacerbate Lebanon’s economic difficulties. However, if the Lebanese government is willing to go ahead with this approach, it is advised that, based upon the results presented and in order to minimise the adverse effects of this policy, the government should adopt a gradual approach because it leads to much less macroeconomic volatility. Another important conclusion is that the government in Lebanon should be aware that the reduction in government expenditures, in order to reduce the budget deficit, is not the best strategy and especially the policy of reduction in government capital expenditure. If the government in Lebanon decides to implement the policy of expansionary monetary policy in order to reduce the budget deficit, our results show that this policy will have some positive effect on Lebanon’s economy, but the government has to be aware that this policy has an inflationary effect.

02Whole.pdf (2221 kB)