The Impact of Macroeconomic Policies on Poverty and Income Distribution: Macro–Micro Evaluation Techniques and Tools edited by François Bourguignon, Maurizio Bussolo, and Luiz A. Pereira da Silva , Houndmills , World Bank and Palgrave Macmillan , 2008 , xviii + 338 pp.
This book comprehensively explores methodologies and techniques of empirical economic models to evaluate the impact on poverty of macroeconomic policies. Traditional aggregate macroeconomic models use the simplest and most restrictive assumption that economic policies do not affect the distribution of welfare. Even micropolicies (policies targeted at specific population groups), when scaled up, are likely to have macro consequences. In order to relax this restrictive assumption, the present volume attempts to link a multi-sector model (computable general equilibrium models: CGE) at the macro level with a microeconomic model. There are many advantages in the recent and more sophisticated group of techniques for linking macro and micro modeling. For example, a macro–micro approach enables different questions to be asked about the poverty and distribution consequences related to policy changes, and investigates the second-round effects of policy changes. The approach also allows a detailed picture of the welfare impacts to emerge, thus enabling a more informed discussion of the social protection policy response to policy changes at the macro level.
The book begins by highlighting the limitations of macroeconomic models that use an assumption of representative household groups (RHG) to link macroeconomic policies and microeconomic data. It then moves to more complex, top-down modeling frameworks, which combine (top) macro models and (down) micro-simulation models. The micro models for simulation can be simple micro-accounting models or behavioral micro models. The book also explores integrated models, in which the macro and micro parts are either linked by iterative feedback loops or solved simultaneously as a single model. The modeling procedures are as follows: (i) the macro (top) model is solved; and (ii) its solution in terms of a vector of aggregate prices, wages, and employment variables – the linking aggregate variables – is used to (a) shock a micro household–level data set (in this case, the micro-simulation is quite simple and broadly corresponds to the micro-accounting incidence analysis, i.e., households [and individuals] do not respond to the price shocks [coming from the top model]); or (b) target the aggregate solution values of a microeconomic model (bottom).
Chapter 2 studies the household welfare impacts of the relative price changes induced by specific trade policy reform scenarios for cereals in Morocco. In this chapter, the price changes induced by the trade policy change are simulated from a general equilibrium analysis done for a joint study by the government of Morocco and the World Bank. The household welfare impacts are analyzed with data from the Moroccan Living Standard Survey. The predicted price impacts from the CGE model are given as for the analysis of household-level impacts. A measure of the monetary value of the gain for a household is obtained by adding the proportionate changes in all prices, weighted by their corresponding expenditure or revenue shares. Having estimated the impacts at the household level, the authors examine their welfare variation before and after the reform, and what impact the reform has on poverty and inequality. According to the results, there is an increase in overall poverty from complete removal of protection, and the impact on poverty is found almost entirely in rural areas.
Chapter 3 describes a framework linking a global CGE model with household survey data and relies on it to estimate the effects of multilateral and regional trade reforms on poverty in four major Latin American economies: Brazil, Chile, Colombia, and Mexico. The results show that, because of different initial positions in terms of economic structure, poverty levels, and trade protection, the poverty effects are quite dissimilar across the four countries studied. Furthermore, the distributional effects of trade are more important than the growth effects.
Chapter 4 presents a new approach that can be used to quantify the effects of macroeconomic shocks on poverty and inequality by overcoming difficulties with multi-sector models with representative household groups. The new approach is needed because the majority of existing multi-sector, multi-household models focus on the inequality between representative groups, and tend to ignore the fact that a majority of households in developing countries generate income from various sources (salaried employment in the formal sector, wage work in the informal sector, and self-employment). This chapter presents a new approach, which combines a micro-simulation model with a standard multi-sector CGE model. The models are used to simulate the full distributional impacts of the Indonesian financial crisis.
Chapter 5 uses a top-down macro–micro model of the Brazilian economy to examine the impacts of the 1998–99 currency crisis in Brazil on the occupational structure of the labor force and on the distribution of incomes. The authors explore a macro model based on a set of investment, savings, and liquidity money supply (IS–LM) equations estimated econometrically on time series data. This “top” model is linked to a microeconometric simulation model of household income formation, estimated on cross-section data from a household survey at the “bottom.” The authors argue that the top-down macro–micro economic model can actually predict the broad pattern of the incidence of changes in household incomes reasonably well, and much better than the alternative approach.
Chapter 6 presents an original approach to the macro–micro link and an application of the methodology to the simulation of the distributional effects of across-the-board trade liberalization in the Philippines. The approach proposed here is the integrated multi-household (IMH) approach, which appears as the only one based on a rigorous theoretical framework that considers all the observed heterogeneity of the population of households (p.180). The authors examine how the IMH approach works, and compare the approach with alternative approaches, such as the representative household and micro-simulation sequential (MSS) approach (p.179).
Chapter 7 presents a simulation model which integrates a static micro-simulation module of labor supply or income and consumption demand with a static CGE-type macro module, and its application to the case of large poverty alleviation policies in Madagascar. The chapter explores three simulations with the objective of improving the situation of the poor: a direct subsidy on agricultural prices; a workfare program; and an untargeted transfer program.
Chapter 8 explores an application of a general equilibrium occupational-choice model to two sectors in Thailand between 1976 and 1996. The authors show that, without an expansion of the financial intermediation sector, Thailand would have evolved with a much lower growth rate, a higher residual subsistence sector, and nonincreasing wages but lower inequality. The financial liberalization resulted in welfare gains and losses to different subsets of the population with a limited impact of foreign capital on growth or the distribution of observed income.
Chapter 9 presents Maquette for MDG simulations (MAMS, p.287), which was produced within a research program on the Millennium Development Goals (MDGs) conducted at the World Bank. MAMS is an economy-wide framework designed to analyze the interaction between the delivery of human development services (health, education, water, and sanitation), the MDGs, growth, and foreign aid. The changing returns to scale in MDGs' targets relating to service delivery are represented by a logistic curve, showing increasing returns to scale at low levels of development indicators and decreasing returns to scale at high level of development indicators. The features that distinguish MAMS from other CGE models are the feedback from and links between different MDGs and the rest of the economy. The chapter describes an application of the MAMS model to strategies related to efforts to achieve the MDGs.
Finally, Chapter 10 highlights some of difficulties to be solved. The first difficulty is data quality, and the second is better modeling of growth or dynamics of economic systems. In addition, the chapter explains frontiers of microeconomic research, such as modeling inequality of opportunity, heterogeneous firms, and the development of structural models that identify clear channels between the quality of governance and economic performance.
The book is carefully written and covers a range of topics in the recent development of macro–micro modeling. In order to deepen the discussion in this book, I would like to provide some comments.
First, the book relates to sophisticated techniques in macro–micro modeling which require an understanding of compilation and solution of traditional CGE models. The book would enhance readers' understanding if it included an introductory chapter explaining elementary procedures of basic CGE models.
Second, as the authors recognize, dynamic aspects of the model still need to be examined. This is because social policy and income transfer concern transfers between generations, and appropriate formulation of saving behavior is needed. Inclusion of methodologies such as generational accounting may be useful.
Finally, because the book explains many classes of alternative formulation of empirical models, a criterion for alternative model comparison needs to be studied. This is because empirical models have to show how well the models fit the data as well as how reasonable the properties of the solutions of the model are. In addition, theoretical foundation and simplicity are also required. For example, Valadkhani's “History of Macroeconometric Modeling: Lessons from Past Experience” (Journal of Policy Modeling 26, no. 2) explains issues regarding model comparison which would be useful for readers of the present book.