Aid, Public Investment, and pro-poor Growth Policies Addis Ababa, August 16-19, 2004 Session 4 An Operational Macroeconomic Framework for Ethiopia Pierre-Richard Agénor Main features. Public capital and production. Allocation of production and prices. Aid allocation. Government budget, BOP, and savings-investment balance. Model equations. Simulations; increase in non-food aid, other shocks (K. El Aynaoui). Main Features
1. Model is fundamentally an aggregate macro framework, with one household, and one (composite) domestic good but imperfectly substitutable to the foreign good; the real exchange rate is thus endogenous (essential to account for the Dutch disease effects of foreign aid). 2. Model accounts for impact of foreign aid on the economy; direct effects on the budget and the BOP, but also on tax effort, public investment expenditure, and public consumption but key distinction between food aid and nonfood aid; food aid increases both domestic demand and supply of goods, whereas non-food aid only is linked to public investment. 3. Model accounts for composition of public investment and capital in health, infrastructure, and education with all components possibly subject to congestion costs: health (population), infrastructure (output, population) and education (stock of raw labor). 4. Model has a simplified labor market; it distinguishes between raw labor and educated labor and accounts endogenously for the process through which education (a public good) is provided. Basic assumption: labor needs to receive some education to be used productively Arguments: simplifies the specification of wage formation; no need to distinguish between skill categories if education expenditure is not disaggregated as well; reduces data requirements.
5. Model accounts for limits on aid absorptive capacity by introducing a non-linearity in the relationship between non-food aid and total public investment with foreign assistance being positively related to public capital outlays up to a certain level of aid, and negatively related thereafter. Depending on parameter estimates, aid may entail significant diminishing returns. 6. The model is linked to a household survey to calculate poverty effects, using the standard IMMPA approach proposed by Agénor, Izquierdo, and Fofack (2003). Steps: 1. From an existing household survey, extract the value of consumption for each household, and given the poverty line and calculate the initial poverty rate, using various standard poverty indicators. 2. Following a policy or exogenous shock, generate the growth rate in per capita consumption of the representative household in the macro model, up to the end of the simulation horizon (say, N periods). 3. Apply this growth rate to the consumption expenditure data for each household in the survey. This gives new consumption levels for each household in the survey, for periods 1,...N.
4. Update the poverty line in the survey by using the growth rate of the composite price index generated by the macro model (implicit assumption: poverty line is constant in real terms). 5. Using the new data on nominal consumption per household and the poverty line, calculate post-shock poverty indicators. Compare with initial indicators to assess the poverty effect of the shock. Implicit assumption: growth is distribution neutral. 7. The model is dynamic; this allows the analysis of dynamic trade-offs that poverty-reduction strategies may entail regarding the impact of policy reforms e.g. between the short-run impact of higher public spending on education and health (on the budget and aggregate demand) and the long-run effects on the productivity and supply of educated labor. 8. The model requires econometric estimation of only a few equations (5-6) others can be calibrated using typical parameters in the CGE literature, and standard national accounts, fiscal and BOP data. Relative ease of use (coded in Eviews), which should ensure replicability. Congestion effects, however, are difficult to measure with any accuracy at this stage (not much empirical work available).
Public Capital and Production Public Capital and Production Public capital in education Public capital in health Congestion Educated labor Congestion Production Population Congestion Public capital in infrastructure Raw labor Effective labor Private capital
Allocation of Production and Prices Allocation of Production and Prices Domestic sales Production Exports Export prices Domestic supply Domestic prices Imports Factor income Food aid Domestic demand Aid Allocation
Aid Allocation Foreign aid Food aid Domestic supply Domestic Production Nonfood aid Public investment Public capital Domestic prices Tax revenue Disposable income Government spending Domestic demand Government Budget Government Budget-Closure I Public investment Lump-sum transfers Public foreign borrowing Tax revenues Overall government deficit Interest payments on foreign debt Aggregate demand Saving-investment balance Consumption investment Private disposable income Non-interest expenditure
Government Budget-Closure II Public investment Lump-sum transfers Non-interest expenditure Tax revenues Overall government deficit Interest payments on foreign debt Aggregate demand Public foreign borrowing Consumption investment Private disposable income Balance of Payments Balance of Payments Imports Exports Current account Private unrequited transfers Debt service Foreign Aid World interest rates Total foreign debt Private foreign borrowing Nominal exchange rate Public foreign borrowing
Savings-Investment Balance Saving-Investment Balance (Closure I) Net foreign savings Overall government deficit Private Saving Saving-investment balance Private consumption, propensity to save, or private investment Domestic demand Model Equations
Potential adverse impact of aid on domestic revenue mobilization effort (see Session 1) aid may create a disincentive for policymakers to incur the political costs of a strengthening in tax administration.
Figure 7 Ethiopia: Composition of Public Fixed Capital Formation (in percent) 120 100 80 60 40 20 0 2001 1999 1997 1995 1993 1991 1989 1987 1985 1983 1981 1979 1977 1975 Health Education Infrastructure Others Source: Government authorities.
Application : Increase in Non-food Aid Baseline: 2003-15, based on recent trends. Closure rule: Ethiopia borrows externally at concessional terms to close its budget gap. Permanent increase in the aid-gdp ratio by 5 percentage points. On the one hand, shock lowers the fiscal deficit (addition to resources), and on the other it raises it, because it increases overall public investment. The initial reduction in the budget deficit gradually disappears. The increase in public investment in infrastructure crowds in private investment and leads to a higher growth rate of about 0.2 percentage points in the long run. Tends to raise the demand for educated labor (factor complementarity). Increase in public investment is allocated across all components (according to initial shares), which therefore implies that the greater demand for educated labor is matched (at least in part) by an increase in supply.
Real GDP at factor cost (% change) Real disposible income per capita (% change) 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Increase in the stock of public capital in health raises the efficiency of labor; increase in public capital in infrastructure raises the marginal productivity of all other production factors (including effective labor). Productivity gains lead to higher output (despite congestion effects), which in turn raises private consumption and lowers poverty. The growth rate of real disposable income per capita rises at its peak by almost 0.6 percentage points.
Poverty Incidence (Deviation from Baseline) 0.00-1.00-2.00-3.00-4.00 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Poverty headcount index (survey data) Disposable income elasticity of -1.0 Important conclusion Potential Dutch disease effect of an increase in aid does not materialize not even in the short run. Because the increase in non-food aid raises public investment, and thus private capital formation investment, the adverse effect of a rise in aggregate demand on prices is offset by the positive supply-side effects of the increase in public and private capital. The poverty rate, based on the survey data, falls by 3.5 percentage points by 2015, whereas the estimate based on a partial elasticity of -1.8 falls by about 4.2 percentage points by 2015. Magnitude of the long-run impact on poverty depends mostly on the strength of supply-side effects which also mitigate Dutch disease effects as a result of a supply response that dominates the increase in aggregate demand resulting from higher government spending.
Other Simulations Changes in level, and composition of foreign aid (the latter implying also change in the level of public investment). Change in the composition of public spending coupled with a reallocation of public investment (for a given level of foreign aid). Calculation of increase in (non-food) foreign aid needed to achieve poverty goal in the MDGs for Ethiopia at horizon 2015, and temporary big push. Extensions
1. More detailed treatment of the labor market (skilled and unskilled labor, to endogenize the wage premium) Problem: treatment of the demand for education (requires an additional market to determine the price of education). 2. Disaggregate non-interest current expenditure to distinguish between teachers salaries and maintenance (or recurrent) spending associated with health and infrastructure. First extension would allow to create a public education input by combining it with public capital in education. It would allow the analysis of the impact of teachers pay (if the quality of teaching is a function of relative wages). Second extension would show that higher investment does put pressure on the budget by increasing future expenditure. 3. To account for distributional issues and some degree of sectoral disaggregation (rural/urban, formal/informal): integrate aid specifications in Mini-IMMPA. Requires significant time investment (need to build a SAM, greater data requirements). 4. Other modifications (see Niger application): direct/indirect taxes; public employment and public sector wage bill.