Dynamic Macroeconomic Models in Emerging Market Economies

This chapter briefly reviews the development of macroeconomics, with emphasis on the dynamic macroeconomic models, DSGE models, used in this book and the corresponding theoretical framework, the NCM.

In this chapter, ways to improve DSGE models are demonstrated and discussed, with emphasis on incentives to introduce financial and housing markets and social stratification into existing dynamic macroeconomic systems. Along with the growing importance of emerging market economies in the global economy, the necessity to develop macroeconomic models for these economies enhances. Aiming to better capture the features of emerging market economies in our models, social and economic characteristics of these economies are summarized in this chapter as well, focusing on three major developing countries—Brazil, India and China.

In this chapter, we derive the traditional dynamic model (RBC/DSGE model) from its origin—the Solow model. This chapter consists of two frameworks: the Solow model and the stochastic (RBC) models. Moreover, early attempts to introduce money into the RBC/DSGE models are demonstrated in the third section of this chapter.

In the first part of this book, we conclude that modern mainstream macroeconomics model is deeply rooted in the NCM theoretical framework. Thus, it is widely cited as the NCM/DSGE model. In this chapter, we derive the fundamentals of the NCM/DSGE model with greater details. The major features of these models, such as intertemporal utility maximization, short-term nominal rigidities and heterogeneities in the production sector, are maintained in the new model we are going to build, as shown in the following chapters. Models derived in the previous chapter are based on the assumption that each market is completely competitive, meaning that there is no nominal rigidity in the economy. Under such a condition, price responds instantly to any innovation, driving the economy back to its equilibrium. Such strong assumption limits the reach and accuracy of these models in macroeconomic analysis. To improve, economists introduce price rigidities into the original RBC/DSGE models. As a result, the NCM/DSGE models emerged. The mechanism that is often referred to as the staggered pricing has become one of the standard mechanisms in the NCM/DSGE models. This chapter consists of three sections. In the first section, we derive the economic behaviours of the manufacturers, including final goods packer and intermediate goods producers. The Calvo pricing mechanism that gives rise to the price rigidity is discussed. The second section examines the household sector, and the general equilibrium is discussed in the third section.

Modern dynamic macroeconomic models are not flawless, they suffer from certain significant limitations. But they provide a highly flexible platform, on which improvements and developments can be realized. As we have discussed, modern mainstream macroeconomics incorporates major contributions of economic research over the last decades, and its empirical DSGE model is of great versatility and flexibility. Theoretical developments of economics can be empirically realized in the corresponding DSGE model by modifying the underlying assumptions and adding more economic features. These qualities are highly desirable for scholars, who attempt to build new dynamic models with more comprehensive structure. In this chapter, we demonstrate an overview of the dynamic macroeconomic system designed for emerging market economies. In the first section, the incentives to develop such a model is discussed. Section 5.2 summarizes the general assumptions underlying the new dynamic macroeconomic model.

This chapter takes the first step to develop the FHSAM. It proposes a dynamic model with two social classes in the economy. This basic two-layer model gives the preliminary attempt to model household heterogeneities in an RBC/DSGE framework. In the first section, we demonstrate the theoretical framework of this model and derive the corresponding mathematical expressions. The general equilibrium and the steady-state condition are calculated after the linearization of the dynamic system. The theoretical summary and testable hypotheses of this model is included in the second section of this chapter.

Based on the basic model, the advanced model is proposed in this chapter. It is advanced in that it further explores the heterogeneities within urban and rural household groups: urban households are divided into lenders and borrowers. Under such an assumption, urban lenders own the capital and lend to other urban households, the urban household borrowers. Both household borrowers and lenders possess housing property, which is explicitly introduced into the utility functions of these households. In contrast to the urban households, the rural ones are assumed to own no housing property, or no real estate with material market value. Such an assumption reconciles the structure of model economy with the facts of situations in emerging market economies. Similar to the previous chapter, the model framework is included in the first section, followed by a summary of theoretical hypotheses.

In this chapter, the theoretical framework of the fully fledged FHSAM is finalized. It maintains certain basic assumptions in the advanced model, such as household heterogeneities (social stratification and/or classification) and nominal rigidity in the non-housing market, and improves the structure of the model economy by installing the financial market. On the supply side, two markets are contained in the full model: the goods market (the non-housing market producing consumption goods, intermediate goods and business investment; the housing market producing real estate for household borrowers and household lenders. In addition, the full model contains an explicit financial market linking household deposits (the bank lending supply) and the firms calling for loans to pay their working capital expenditure (bank lending demand). The theoretical summary and hypotheses of the fully FHSAM are discussed in the second section of this chapter.

In the previous chapters, we have derived the FHSAM framework, which is stochastic and dynamic with complex mathematical representation. The analytical solution, if there is one, to models of this kind is extremely difficult to find. But thanks to the development of modern dynamic optimization theory and to the advancement in mathematics, we can better examine the characteristics of DSGE models using scientific computation and perturbation methods. In almost all empirical modelling exercises, perturbation is used as the standard method to solve problems of the DSGE type of models. Adopting this method, we need to first find the steady state and then to analyse the states that deviate from the steady state. In general, linearization, including first-order and second-order Taylor expansions, is conducted to study the approximation near the steady state (the latest version of Dynare software package supports third-order Taylor expansion, but the mainstream analysis focuses on the second-order Taylor expansion). In this chapter, we discuss the mathematical methods to solve a dynamic economic system. Several major components of mainstream tool such as linearization, state-space representation, and the Blanchard–Kahn condition, are discussed in the following sections.

The first step of empirical analysis is model identification, assigning appropriate values to all the parameters included in the model. The importance of model identification should be emphasized in DSGE modelling exercises because models of this type are typically represented by a large number of equations and parameters, each with explicit economic meaning. The theoretical framework is crucial, but it is only as good as its empirical model that is decently identified. Though different in applications, the methodologies adopted in most of the DSGE literature fall into two general categories: calibration and estimation, both of which are used in this book. This chapter proceeds as follows: we summarize the fundamentals and procedures of parameter calibration and estimation in the first two sections respectively; the third section discusses the software packages used in this book.

As the input to empirical analysis, data needs to be carefully collected. In this chapter, we summarize the data samples observed in Brazil, India and China. Time series of both real and financial sectors are collected. Due to data accessibility in BIC, the time frequency of the data set is quarterly. Statistics and stylized facts of the data set are demonstrated in this chapter as well.

This chapter finalizes the empirical analysis of the FHSAM. All three models, the basic model, the advanced model and the full model have been empirically analysed, using data collected in Brazil, India and China. Both calibration and estimation methodologies explained in the previous chapter are adopted to identify model parameters. The statistical features of the empirical results are compared with stylized facts, indicating good explanatory power of the FHSAM framework in BIC. Generally, this chapter contains three sections, each examines the empirical analysis of the corresponding model.

This chapter concludes the findings of the FHSAM, emphasizing on the comparison of economic features between model economy, the actual condition and theoretical hypotheses. The contribution of the FHSAM to policymaking scheme in BIC is also discussed in this chapter. Moreover, the discussion on the potentials of the FHSAM is included in this chapter.