یک مدل اقتصاد باز کوچک با هزینه های معامله در سرمایه های خارجی

In this paper we develop a stochastic dynamic general equilibrium model for a small open economy in the real business cycle modeling tradition. Household preferences depend on private and public consumption and leisure. Production technology depends on public capital. Government finances its investment, consumption and transfer payments by means of a proportional income tax rate. Households buy and sell foreign assets in an international capital market with transaction costs and also receive transfer payments from abroad. We calibrate the model for the Greek economy. The volatility, persistence, and co-movement properties of the business cycle component of the data generated by the model are broadly consistent with the actual behavior of the corresponding actual data. We use the model to investigate the response of major macroeconomic variables to temporary and permanent changes in government policy variables, foreign transfers, and the rate on return of foreign assets. We find that increasing government consumption and domestic transfers financed through distorting taxation lowers capital, labor (in most cases) and output while it increases foreign asset holdings, both in the short and the long run. Increasing government investment financed by distorting taxation eventually increases capital and output and decreases labor. Finally transfers from abroad decrease capital, labor and output, while they increase consumption.

There are serious doubts whether simple Real Business Cycle (RBC) models can be helpful in explaining the behavior of economies other than that of the United States.1 These doubts are justified on the presumption that most of these economies have features that are very difficult to capture in simple RBC setups (see, e.g., Prescott, 1986). Notably, it is argued that most economies other than that of the United States should be better modeled as `small open economies' and, in several cases, with foreign capital market imperfections (see, e.g., Mendoza, 1991, p. 815). Further, these doubts are justified on the ground that, in most of these other economies, government intervention plays an important role (see, e.g., Gali, 1994). Furthermore, the overwhelming evidence that the labor market is the most differentiated institutional feature in the economies of the developed world should be also considered (see, e.g., Danthine and Donaldson, 1993, pp. 29–30; Fiorito and Kollintzas, 1994, pp. 259–260; Christodoulakis et al., 1995). Thus, with a few notable exceptions, the success of modeling the United States Economy via RBC models has not been accompanied by models for other countries.2 This paper is a further attempt to remedy this omission. In particular, in this paper we extend the standard RBC model (Prescott, 1986) to study the behavior of a small open economy that is characterized by impediments to its access to the foreign capital markets and a relatively large and distorting public sector. Our second objective is to calibrate this model on data for the Greek economy from 1960 to 1992 and look first at the descriptive power of the model and second to investigate the response of major macroeconomic variables to temporary and permanent changes in government policy variables. We are especially interested in studying the effects of foreign transfers, given that Greece has been the `beneficiary' of a relatively big transfer program from the other EU countries since 1989 (the so-called `Delors I and II packages'). In so doing, we ignore problems stemming from the fact that labor markets in Greece are highly centralized.3 Moreover, the lack of appropriate and/or reliable data is a problem for Greece. However, we managed to avoid several of these problems. In particular, the most severe problem that has to do with the public finance accounting we tried to ameliorate it by a particular modeling assumption which will be stressed in Section 2.4. Household preferences depend on private and public consumption and leisure as in Christiano and Eichenbaum (1992). Government finances its consumption and makes investment in capital, engages in transfer payments to the private sector by way of a proportional income tax. Thus, in this model, as in Baxter and King (1992), Gali (1994) and Jonsson and Klein (1996), government policy changes are not neutral since they affect the effective income tax rate introducing incentive effects for capital accumulation and work effort. An international capital market exists where households can buy and sell foreign assets. Households also receive transfer payments from abroad. Motivated by the various capital flow controls that characterized the Greek economy until 1992 (see, e.g., Alogoskoufis, 1995; Christodoulakis and Karamouzis, 1995), we introduce transaction costs in the foreign sector. Thus, one of the questions we are interested in is whether the increase in government consumption and domestic transfers payments and the decrease in government investment along with the increased foreign transfers that characterized the economy over the last twenty years affected output, consumption, investment, work, etc. Other than the application to the Greek economy, our contribution lies with the fact that we extend the existing literature on small open economy RBC models (e.g., Mendoza, 1991; Correia et al., 1995; Bruno and Portier, 1995) by introducing transaction costs in the foreign capital markets, while simultaneously introducing stochastic government behavior as in Gali (1994), Jonsson and Klein (1996) and McGrattan et al. (1997). We find that our model's descriptive power should be considered very good, both with respect to a comparison between the actual and simulated data, as well as relative to the performance of a simple (Prescott (1986)-type) closed economy model. We also find that increasing government consumption and domestic transfers financed through distorting taxation, lowers capital, labor (in most cases) and output while it increases foreign asset holdings, both in the short and the long run. Increasing government investment financed by distorting taxation, eventually increases capital and output and decreases labor. Finally, transfers from abroad decrease capital, labor and output, while they increase consumption. The plan of the paper is as follows: Section 2presents the model. Section 3defines and solves for the competitive equilibrium and characterizes the steady state growth path. Section 4calibrates the model and looks at the descriptive power of the simulated data. Section 5discusses the results from the investigation of the response of major macroeconomic variables to temporary and permanent changes in government policy, foreign transfers, and return on foreign assets. Section 6concludes the paper.

In this paper we develop a stochastic dynamic general equilibrium model with transaction costs on net foreign asset accumulation and a distortionary public sector that is capable of explaining several of the stylized facts of the Greek business cycle. Household preferences depend on private and public consumption and leisure. Government makes investment in capital, engages in government consumption, and provides transfer payments to the private sector while maintaining a balanced budget. An international capital market exists where households can buy and sell foreign bonds and receive net transfers from abroad. The volatility, persistence and comovement properties of the data generated by the model are broadly consistent with the actual behavior of the Greek Economy, from 1960–1992. We also use the model to investigate the response of major macroeconomic variables to temporary and permanent changes in government policy variables, foreign transfers, and the international interest rate. Our simple model does quite well in reproducing several of the key stylized facts of the Greek Business Cycle. The only exception is the labor market, which for reasons already mentioned, does not seem to follow well the Walrasian framework. Clearly, more work is needed towards this issue. It should be interesting to integrate the RBC methodology with the dynamic firm-union bargain set-up (see Eberwein and Kollintzas, 1995). One other source of the still existing problems is due to the fact the foreign transfers were assumed (as a tractable device to solve the model) proportional to GDP which are certainly not. One of the surprising properties of the model is that it is consistent with what some believe to be an idiosyncrasy of the Greek Economy – the fact that increases in foreign transfers leads to less output and work and more consumption and foreign investments. The effects of both temporary and permanent increases to total factor productivity are the typical responses found in RBC models. The response of an increase in the GDP share of foreign transfers increases consumption, foreign assets and the trade balance and decreases work, investment output and both private and public capital. The main effect of an increase in pf (i.e., decline in r*) is to reduce foreign asset holdings and the trade balance and increase domestic private investment. Our model predicts that increases in the GDP share of government consumption have adverse effects on output and the factors of production as well as all other variables. From the impulse response function analysis, we find that a 1% permanent increase in the GDP share of government consumption leads to a 4% to 10% fall in output (for ϑ=0, output falls by 7.8%). Increases in the GDP share of government investment increase output and all kinds of capital and decrease labor. The response of an increase in the GDP share of domestic transfers is qualitatively similar but quantitatively smaller, to an increase in the GDP share of government consumption. Thus our model suggests that the increases in the shares of government consumption, and foreign and domestic transfers over the last 20 years have worked to reduce the performance of the Greek economy both with respect to its steady state and its transition to it (see Kollintzas and Vassilatos, 1996). The primary mechanism behind this result is the distortions on the incentives to save and work. Our model's prediction for the volatility, persistence and comovement properties of the key macroeconomic variables as well as the shape of the impulse response functions were insensitive to the value of ϑ. The role of government consumption in preferences affects only the quantitative behavior of the impulse response functions. The transition towards the steady state was captured by choosing the appropriate initial values. In future work it would be interesting to calibrate the economy in two periods (before and after 1974) thus achieving a closer fit with real data characteristics. Presumably, the model would be able to better account for most of these stylized facts if we calibrated the model in two steady states, each one being associated with a different set of values for the tax policy variables and then follow the usual procedure.