Consumption and macroeconomic policies
Evidence of asymmetry in developing countries
The Authors
Magda Kandil, International Monetary Fund, Washington, DC, USA
Ida Aghdas Mirzaie, Department of Economics, The Ohio State University, Columbus, Ohio, USA
Abstract
Purpose – The paper aims to examine asymmetry in the cyclical behavior of private consumption in a sample of nine developing countries in the Middle East.
Design/methodology/approach – The empirical model includes three policy variables: government spending, the money supply, and the exchange rate. Anticipated movements in these variables are likely to vary with agents' forecasts of macroeconomic fundamentals and, therefore, determine planned consumption. Unanticipated policy changes, in contrast, determine cyclical consumption.
Findings – The results indicate that fluctuations in private consumption are mostly cyclical. The stabilizing function of policy shocks varies across countries and appears to be asymmetric within countries.
Originality/value – Asymmetry necessitates a thorough evaluation of the positive and negative effects attributed to changes in policy variables and the necessary reforms to relax binding constraints.
Article Type:
Research paper
Keyword(s):
Consumption; Developing countries; Economic cycles; Economic stability; Middle East.
Journal:
International Journal of Development Issues
Volume:
6
Number:
2
Year:
2007
pp:
83-105
Copyright ©
Emerald Group Publishing Limited
ISSN:
1446-8956
Introduction
Consumption spending is a major component of aggregate demand. Table I presents the shares of private consumption to GDP for a sample of developing countries in the Middle East in 2000 and the average over a longer sample period. The share of consumption to GDP fluctuated during the 1990s. Fluctuations in consumption determine fluctuations in the saving rate and, in turn, the saving/investment balance. The latter is the mirror image of the current account balance. An increase in consumption eats up national income, necessitating continued reliance on foreign resources to finance domestic investment, i.e. an increase in the current account deficit. Given fluctuations in the domestic resource gap, it is worthwhile to study the behavior of private consumption using available data for a sample of developing countries that have never been studied for this purpose before.
Why is private consumption not growing smoothly with nominal income over time? Fluctuations in the marginal propensity to consume indicate unstable plans for consumption, which is likely to have adverse effects on growth and inflation, further contributing to aggregate uncertainty. It is also worth noting that the marginal propensity to save is, generally, high in developing countries. Such a high saving rate indicates fluctuations in consumers' confidence with economic outlook in light of aggregate uncertainty. A high unstable saving rate raises concerns about the adverse effects of unstable consumption on economic growth and warrants a thorough investigation of determinants of consumption spending.
A positive cyclical response of private consumption to real growth is desirable. It indicates a reduction in private consumption with a decline in real growth. This positive correlation reduces pressure on price inflation given constraints on the supply side. Further, the positive response indicates the need for policy intervention to stimulate economic conditions during periods of slowdown. It is, therefore, necessary to evaluate developments in private consumption with major policy tools: fiscal policy, monetary policy, and exchange rate policy.
In a previous investigation (Kandil and Mirzaie (2007), consumption varies procyclically with real growth. Moreover, consumption decreases in response to a higher interest rate and/or price inflation. In this paper, we extend our earlier research to study the role of stabilization policies in determining planned and cyclical consumption. Using a rational expectation model, we decompose policy variables into anticipated and unanticipated components. This decomposition aims at separating fluctuations in planned consumption in the face of anticipated forecasts compared to cyclical consumption that varies in the face of unanticipated shifts. The anticipated component varies with agents' forecasts of macroeconomic fundamentals. By construction, the steady-state component of policy variables varies with available information at the time of forming forecasts. In contrast, the cyclical component of consumption varies with random uncertainty impinging on the economic system which is the domain of unanticipated changes in policy variables beyond agents' forecasts.
We will study how consumption spending varies with stabilization policies, including both fiscal and monetary policies, in a sample of Middle Eastern countries
Anticipated changes in policy variables are likely to determine the steady state income. We will investigate whether consumption moves with anticipated changes in the money supply, government spending, and the exchange rate. In addition, one expects that transitory shocks in policy variables will determine cyclical consumption. Moreover, policy shocks are decomposed into expansionary and contractionary shocks. Using this decomposition, we study asymmetry in the cyclical behavior of consumption in the face of policy shocks.
Countries included in the study are Algeria, Egypt, Iran, Jordan, Libya, Oman, Pakistan, Syria, and Tunisia. The sample period of investigation varies based on data availability. The investigation will evaluate the final findings from a regional standpoint to shed light on the cross-regional similarities and the role of policy makers in determining and stabilizing consumption in a sample of developing countries.
The remainder of the paper is organized as follows. Next section provides a literature review. Followed by a section that introduces a theoretical background to model private consumption. It is followed by two sections, one outlines the empirical models, and the other describes the econometric methodology, respectively. The penultimate section presents empirical results. The summary and conclusion are presented in the last section.
Literature review
Theoretical developments have incorporated rational expectations to formulate agents' strategy regarding consumption in relation to permanent income. Empirical research has, however, challenged the steady nature of consumption in relation to permanent income. Hall (1978) indicates that consumption expenditure follows a random walk, which implies that only unexpected policy changes can affect consumption, given knowledge of last period's consumption
Heller and Starr (1979), Falvin (1981), Hall and Mishkin (1982), and Campbell and Mankiw (1989) present empirical findings that suggest the rational expectation permanent income hypothesis does not hold because consumption displays an excess dependence on current income. Campbell and Mankiw (1989, 1990, 1991) show that for nondurable and service consumption, the elasticity of intertemporal substitution is basically zero.
It is expected, however, that the excess dependence of consumption on current income is more pronounced in developing societies due to capital imperfection and credit rationing. As Kaynak et al. (1995) state, people tend to use more cash in developing countries rather than using credits. Yet, for these countries, the link between consumption and the real interest rate may exist for durable consumption. If true consumers may indeed adjust their consumption allocations in response to changes in real interest rates.
In addition, the consumption level in Middle Eastern countries maybe influenced by the real value of their currency via the effect on import prices and price inflation. According to Diaz-Alejandro (1963), devaluation may lead to a reduction in consumption and real income. He uses a three-good model of exportable, importable, and home goods with relatively price-inelastic exports and imports and consumption functions displaying higher saving propensities for non-wage earners. The underlying mechanisms behind the devaluation impact on consumption lies in the transfer of real income from workers to producers who have higher propensities to save out of factor income. Krugman and Taylor (1978) and Barbone and Rivera-Batiz (1987) have formalized the same view. The Krugman-Taylor paper formalizes several channels of contractionary influences, which are likely to prevail, particularly in developing countries. Krugman and Taylor discuss that devaluation will lead to a reduction in real output if devaluation raises prices of traded goods relative to home goods, giving rise to windfall profits in export- and import-competing industries. If money wages lag the price increase and if the marginal propensity to save from profits is higher than from wages, national savings will go up, and consumption and real output consequently will decrease.
Our approach to the analysis of private consumption builds on our earlier work (Kandil and Mirzaie, 2006). We extend the investigation and employ rational expectations to separate private consumption spending into a planned component that varies with anticipated policy variables and a cyclical component that varies with surprises, i.e. unanticipated policy shocks. The objective is to study the role of stabilizing policies in accommodating and/or stabilizing cyclical consumption, as well as in determining planned consumption.
Theoretical background
In the real world, stochastic uncertainty may arise in the macroeconomy. Economic agents are assumed to be rational. Accordingly, rational expectations distinguish planned behavior from cyclical behavior in the face of random transitory fluctuations. Rational agents engage in a process of forecasting macroeconomic fundamentals. This forecast represents the steady state trend of the macroeconomy. Random fluctuations are assumed to be realized around this steady state trend.
The demand side of the economy is specified using standard IS-LM equations with a modification for an open economy. The specifications below describe the demand and supply sides of the macroeconomy. The subscript t denotes the current value of the variable: Equation 1 Equation 2 Equation 3 Equation 4 Equation 5 Equation 6 Equation 7 Equation 8 Equation 9 Equation 10 In equation (1), real consumption expenditure, c, varies positively with real disposable income, y
d. Consumption varies negatively with the interest rate, int
t
. An increase in rer is consistent with currency appreciation. An appreciation of the real exchange rate decreases the cost of tradables and, therefore, consumption of non-tradables. The combined effect will depend on the elasticity of consumers' substitution between tradables and non-tradables, both embedded in c
t
. In developing countries, this elasticity is likely to be small, reinforcing the positive effect of currency appreciation on consumption. In equation (2), the nominal interest rate is the sum of the real interest rate, r, and inflationary expectation, (E
t
p
t+1−p
t
), where p
t
is the aggregate price level and E
t
is agents' forecast conditional on information available at time t
In equation (5), real investment expenditure, i, varies negatively with the real interest rate
Substituting all equations into the equilibrium condition for the goods market results in the expression for real income. It is a function of the exchange rate, the domestic price level, the foreign price level, and the domestic interest rate. This expression is the IS equation which describes the negative relationship between real income and the real interest rate.
In equation (9), equilibrium in the money market is obtained by equating the demand and supply of real money balances. The real money supply is determined by nominal balances, m, deflated by price, p. The demand for real money balances is positively related to real income and inversely related to the nominal interest rate.
The LM equation is determined by the equilibrium condition in the money market. It establishes a positive relationship between real income and the real interest rate. Solving for the interest rate, r, from the LM equation and substituting the result into the IS equation results in the equation for aggregate demand.
Aggregate supply in equation (10) varies with determinants of the output supply in the production function, s 0, and output price surprises, (p t −E t−1 p t ). Rational agents forecast the aggregate price level conditional on information available at time t−1, E t−1 p t . Aggregate demand shifts induce changes in the output price around its forecasted value. These fluctuations create output price surprises that are the domain of random unexpected shocks on the demand and supply sides of the economy.
The solution of the model follows the details in Kandil and Mirzaie (2003). In brief, combining aggregate demand and aggregate supply, we solve for output price surprises as a function of monetary surprises, (m t −E t−1 m t ), government spending surprises, (g t −E t−1 g t ) and exchange rate surprises (rer t −E t−1rer t ). Substituting the solution for output price surprises into the output supply, we solve for y t . Substituting the solution for y t into the aggregate demand equation, we solve for the price level which varies with anticipated and unanticipated policy variables.
Anticipated changes are the domain of anticipated shifts in the economy that vary with agents' forecasts of macroeconomic fundamentals, primarily policy shifts. Adjustments in private consumption with respect to anticipated policy changes identify the long-term path of consumption. Unanticipated changes are deviations around these forecasts that vary with random shocks impinging on the economic system. Adjustments in private consumption with respect to unanticipated policy changes identify short-term fluctuations in consumption.
Having solved for disposable income and the interest rate, we solve for private consumption. According to this solution, planned private consumption varies with anticipated changes in policy variables, government spending and the money supply, as well as anticipated changes in the exchange rate. Cyclical consumption varies unexpectedly with unanticipated changes in these variables. The solution of consumption can then be approximated as follows: Equation 11 An increase in government spending increases income and the interest rate. The effects of an increase in government spending on private consumption is likely to depend on the marginal propensity to consume out of income, c1, and the sensitivity of private consumption to a change in the interest rate, c3, as well as on the relative effects of government spending on income and the interest rate. If the income channel dominates an increase in government spending increases disposable income and private consumption. In contrast, if the interest rate channel dominates, an increase in government spending increases the interest rate and decreases private consumption. The positive and negative channels apply to both anticipated and unanticipated growth in government spending. The distinction will differentiate planned from cyclical fluctuations in consumption with respect to changes in government spending.
An increase in the money supply increases income and decreases the real interest rate (the liquidity effect). Theoretical developments have established, however, the positive effect of an increase in the money supply on inflationary expectations (the Fisher effect). If the Fisher effect dominates the liquidity effect, the nominal interest rate may be rising following an increase in the money supply. The effect of an increase in the money supply on private consumption will depend, therefore, on the marginal propensity to consume out of income, c1, the sensitivity of private consumption to a change in the interest rate, c3, and the relative effects of the money supply on income and the interest rate. The positive and negative channels apply to both anticipated and unanticipated growth in the money supply. The distinction will differentiate planned from cyclical fluctuations in consumption with respect to changes in the money supply.
An appreciation of the real exchange rate decreases the cost of imports. As the demand for imports increases, private consumption of non-tradables decreases. The final effect on consumption will depend on the relative increase in tradable consumption compared to the reduction in non-tradable consumption underlying adjustments in c t . The positive and negative channels apply to both anticipated and unanticipated currency appreciation. The distinction will differentiate planned from cyclical fluctuations in consumption with respect to changes in the exchange rate.
It is clear that a number of channels determine the response of private consumption to policy variables. Given the complexity of these theoretical channels, the response of private consumption to each of the policy variables may prove to be nonlinear. Hence, in the empirical model we decompose shocks to policy variables into positive and negative components to detect possible asymmetric effects on private consumption. We expect the estimated coefficients to vary across countries depending on the relative strengths of underlying channels.
Empirical models
The empirical investigation analyzes annual time-series data of private consumption in nine countries in the Middle East: Algeria, Egypt, Iran, Jordan, Libya, Oman, Pakistan, Syria, and Tunisia. Data for most countries cover 1963-2002. For Algeria and Tunisia, the sample period extends through 2003. For Oman, the sample period is 1967-2002. For data definition and sources, see Appendix
We estimate a reduced-form equation that replicates the solution of private consumption in the model. To account for correlations across policy variables, private consumption varies with government spending, the money supply, and the exchange rate. Each policy variable is decomposed into anticipated and unanticipated components. The response of private consumption to anticipated policy shifts will indicate the degree by which anticipated policy shifts gauge planned consumption decision in the steady state. The response of private consumption to unanticipated policy shocks will indicate the cyclical nature of consumption spending and the success of stabilization policies in countering this cyclicality
Of particular interest is to study asymmetry in consumption fluctuations in the face of policy shocks. To that end, policy shocks are decomposed into positive and negative components. The sign and significance of parameters measuring the response of private consumption to positive and negative shocks will be studied to evaluate the degree of asymmetry.
Having tested for non-stationarity, the empirical model is specified in first-difference form
In theory, consumption varies with anticipated and unanticipated variables. Nonetheless, non-stationarity requires that consumption is detrended by removing a stochastic trend component. The resulting detrended series is, therefore, a function of anticipated and unanticipated components of the detrended explanatory variables that determine private consumption in theory.
Since, the model is estimated in first-difference form, we should test if the non-stationary dependent variable is jointly cointegrated with all non-stationary right-hand side variables. Given evidence of cointegration (see Table AII in the Appendix), the error correction term is included in the empirical model
To establish robustness and draw further insights, we estimate another version of the empirical model allowing for variation in data measures. Nominal consumption varies with nominal government spending, nominal money, and the nominal exchange rate
To study asymmetry, each of the shocks in the empirical models is decomposed into positive and negative components as follows: Equation 13 Shocks to government spending are decomposed into positive and negative components, posg and negg. Expansionary and contractionary shocks to the money supply are denoted by posm and negm. A positive shock to the exchange rate, posr, indicates currency appreciation while a negative shock, negr, indicates currency depreciation.
Econometric methodology
The surprise terms that enter models (12) and (13) are unobservable, necessitating the construction of empirical proxies before estimation takes place. Thus, the empirical models include equations describing agents' forecast of the change in the log values of the exchange rate, the money supply and government spending. All variables are first-differenced to render the series stationary, as described in Table AI.
To decide on variables in the forecast equations, a formal causality test is followed. Each variable is regressed on two of its lags as well as two lags of all variables that enter the model: the change in the log value of income, the interest rate or price, government spending, the money supply, and the exchange rate. The joint significance of the lags is tested for each variable (Table AIII). Accordingly, the forecast equations account for the lags of variables proven to be statistically significant.
Surprises in the empirical model are obtained by subtracting the above forecasts from the actual change in the variable results in surprises in the empirical model. Following the suggestions of Cover (1992), positive and negative shocks underlying surprises are defined for joint estimation as follows: Equation 14 Where, abs(.) is the absolute value operator and shock t measures unanticipated policy surprises.
Given empirical proxies in the model specification, estimating a two-step procedure would suffer from the “Generated regressor problem.” Alternatively, to obtain efficient estimates and ensure correct inferences (i.e. to obtain consistent variance estimates), the empirical models are estimated jointly with a forecast equation for each anticipated regressor, following the suggestions of Pagan (1984, 1986).
To account for endogenous variables, instrumental variables are used in the estimation of the empirical models. The instrument list includes four lags of all variables in the model: price, the interest rate, income, money, government spending, and the exchange rate. In a few cases, the number of lags has been modified until the estimation did converge. The paper's evidence remains robust with respect to modifications that alter variables or the lag length in the forecast equations and/or the instruments list.
Following the suggestions of Engle (1982), the results of the test for serial correlation in simultaneous equation models are consistent with the presence of first-order autoregressive errors. To maintain comparability, it is assumed in all models that the error term follows an AR(1) process. The estimated models are transformed, therefore, to eliminate any possibility of serial correlation. The estimated residuals from the transformed models have zero means and are serially independent.
Empirical results
Table II presents the evidence of estimating real consumption as a function of real government spending, real money, and the real exchange rate.
Planned real consumption does not vary significantly with anticipated real money growth in any country. In contrast, a positive shock to monetary growth stimulates real consumption in Syria. In Syria, monetization is pursued to finance government spending, which provides support for a good share of wages and salaries in the economy. Contractionary shocks to the money supply, in contrast, have insignificant effects on real private consumption in Syria
An expansionary shock to monetary growth decreases real consumption growth in Oman. Expansionary monetary growth decreases the interest rate and stimulates capital outflows. Private consumption also decreases in the face of contractionary shocks to real monetary growth in Oman. The reduction in liquidity decreases available credit.
Monetary shocks appear, therefore, to be asymmetric in determining real consumption spending in Oman, which is supported by a formal statistical test. The interest rate channel dominates in the face of expansionary shocks while the liquidity channel dominates in the face of contractionary shocks.
Planned consumption does not vary significantly with anticipated real government spending in any country. Consistent with the dominant role of the government in providing employment in Syria, expansionary shocks to government spending stimulate private consumption. Contractionary shocks to government spending decrease private consumption significantly in Egypt and Pakistan. This is consistent with the contractionary effect of a reduction in government spending on income. Asymmetry is supported statistically in both cases.
A reduction in government spending stimulates an increase in private consumption in Jordan and Syria. The reduction in government spending increases available credit for private activity. In both cases, asymmetry in the response of private consumption to expansionary and contractionary government spending shocks is supported by the results of a formal statistical test.
Both anticipated and unanticipated currency appreciation has a significant positive effect on private consumption in Egypt. Currency appreciation decreases the cost of imports and increases real income, raising consumption of both tradables and non-tradables. Unanticipated currency depreciation, in contrast, does not have a statistically significant effect on private consumption in Egypt. Asymmetry in the response of private consumption to exchange rate shocks is supported using a formal statistical test.
Currency depreciation decreases real consumption spending significantly in Iran, Jordan, and Oman. Statistical significance supports asymmetry in the response of private consumption to unanticipated currency appreciation and depreciation in Jordan and Oman.
Overall, fluctuations in consumption are mostly transitory in the face of unanticipated policy shocks. The contractionary effect of policy shocks appears more pervasive compared to the expansionary effects on private consumption. The evidence provides limited support to the asymmetric (varying) effects of positive- and negative-policy shocks.
To substantiate the evidence, Table III contains the results of estimating nominal consumption as a function of nominal government spending, nominal money and the nominal exchange rate. Anticipated nominal monetary growth increases private consumption significantly in Jordan
A reduction in monetary growth decreases private consumption significantly in Iran. Monetary growth increases liquidity and, therefore, consumption spending. A reduction in monetary growth increases private consumption significantly in Syria. A reduction in monetary growth coincides with a reduction in government spending and inflation, which has a positive effect on private consumption. As the effect of expansionary monetary shocks is not significant, asymmetry is supported by a formal statistical test.
Anticipated growth in government spending stimulates growth in private consumption in Libya. Unanticipated growth in government spending increases private consumption significantly in Oman. In oil-producing countries, government has a dominant role on economic activity and employment. There is no comparable significant evidence with respect to contractionary shocks. Asymmetry is supported by the results of a formal statistical test.
A negative shock to government spending decreases private consumption significantly in Pakistan. Government spending provides employment and supports wages and salaries. There is no comparable response with respect to expansionary shocks. Asymmetry is not supported statistically.
Planned consumption does not vary significantly with anticipated exchange rate appreciation. Unanticipated appreciation of the exchange rate increases private consumption significantly in Egypt and Libya. Appreciation increases imported consumption and real income. The latter channel stimulates an increase in consumption of non-tradables. In contrast, unanticipated appreciation decreases private consumption significantly in Pakistan. Agents capitalize on currency appreciation by decreasing money demand and consumption of non-tradables.
Unanticipated depreciation increases private consumption significantly in Egypt, Libya, and Syria. Agents switch demand to non-tradables following depreciation. Further, depreciation increases the domestic price of non-tradables and, hence, the nominal value of private consumption. Asymmetry is supported statistically in Egypt, Libya, and Pakistan.
In contrast, unanticipated depreciation increases the cost of imports and decreases private consumption significantly in Jordan. Statistical significance supports asymmetry in the effects of currency fluctuations on private consumption in Jordan.
Overall, fluctuations in nominal consumption are mostly cyclical. The expansionary effects of policy shocks appear to be limited on private consumption. Further, the expansionary effects of policy shocks on private consumption are not matched by equal contractionary effects and vice versa.
Conclusion
The analysis of this paper has focused on a sample consisting of nine developing countries. Theory has distinguished between cyclical and planned fluctuations in private consumption. Economic agents make planned consumption decisions in response to anticipated changes in macroeconomic fundamentals and forecasts of policy variables. In contrast, random transitory fluctuations impinging on the economic system determine cyclical consumption.
In a previous investigation, Kandil and Mirzaie (2006), we report evidence that private consumption in developing countries varies procyclically with real growth. Private consumption decreases in the face of a higher interest rate and/or a higher price inflation. Cyclicality in private consumption with unanticipated macroeconomic developments provides evidence on the level of uncertainty impinging on the economic system in developing countries.
The present investigation aims at identifying the role of macro policies in accommodating or stabilizing cyclical fluctuations in private consumption. We estimate a reduced-form equation that explains private consumption as a function of policy variables: government spending, the money supply, and the exchange rate. Planned consumption varies with anticipated forecasts of policy variables. Cyclical consumption varies with unanticipated policy shifts.
The evidence, in general, indicates that anticipated policy shifts have limited effects on planned consumption. The random component dominates policy shifts, minimizing the significant effect of anticipated policy shifts on private consumption in developing countries.
Cyclical fluctuations in private consumption spending in the face of policy shocks vary across countries. Expansionary monetary policy is significant in stimulating consumption growth in Syria, Pakistan, and Tunisia. Monetary expansion increases liquidity and stimulates consumption growth. In Oman, however, expansionary monetary policy decreases private consumption growth. The reduction in the interest rate in the face of monetary growth increases capital outflows, countering the effects of monetary shocks. Contractionary monetary shocks decrease the growth of real private consumption in Oman and the growth of nominal consumption in Iran. The reduction in liquidity appears to be significant in curbing consumption growth. Nonetheless, the growth of nominal consumption appears to be increasing in the face of contractionary monetary shocks in Syria, signaling ineffectiveness of monetary policy
Consistent with the dominance of government on the economies of developing countries under investigation, expansionary shocks to government spending stimulate private consumption in Syria and Oman. Increased government spending supports higher wages and salaries in the public sector, which provides support for the largest share of employment. Similarly, contractionary shocks to real government spending have a depressing effect on private consumption in Egypt and Pakistan. Nonetheless, the reduction in government spending in Jordan, where a variety of reform measures have been underway, stimulates the growth of private consumption
Unanticipated currency appreciation stimulates the growth of private consumption in Egypt and Libya. Currency appreciation decreases the cost of imports and raises real income, increasing consumption of tradables and non-tradables
Overall, the evidence presents a more important role for the stabilizing function of policy shocks compared with anticipated (steady-state) policy shifts on private consumption
It is worth noting that the model specification does not capture significant fluctuations in consumption in several countries. This provides evidence of policy failure in these countries to determine fluctuations in private consumption, leaving significant randomness induced by endogenous domestic variables (real output, price, and the interest rate, see Kandil and Mirzaie, 2006), and/or exogenous shocks. This evidence should alert policy makers in these countries to the need to design more proactive policies that should target the largest component of aggregate demand, private consumption, and reduce its fluctuations in response to domestic and external shocks impinging on the economic system.
Equation 1
Equation 2
Equation 3
Equation 4
Equation 5
Equation 6
Equation 7
Equation 8
Equation 9
Equation 10
Equation 11
Equation 12
Equation 13
Equation 14
Table IShares of private consumption to GDP
Table IINonlinear 3SLS parameter estimates
Table IIINonlinear 3SLS parameter estimates
Table AIThe Kwiatowski, Phillips, Schmidt, and Shin (KPSS) statistics for null of level stationary
Table AIICointegration test results
Table AIIIThe results of endogeneity tests
References
Barbone, L., Rivera-Batiz, F. (1987), "Foreign capital and the contractionary impact of currency devaluation, with an application to Jamica", Journal of Development Economics, Vol. 26 pp.1-15.
Campbell, J.Y., Mankiw, N.G. (1989), "Consumption and interest rates: reinterpreting the time-series evidence", in Blanchard, O.J., Fisher, S. (Eds),NBER Macroeconomics Annual, MIT Press, Cambridge, MA, .
Campbell, J.Y., Mankiw, N.G. (1990), "Permanent income, current income, and consumption", Journal of Business and Economic Statistics, Vol. 8 No.3, pp.265-79.
Campbell, J.Y., Mankiw, N.G. (1991), "The response of consumption to income, a cross-country investigation", European Economic Review, Vol. 35 pp.723-67.
Cover, J.P. (1992), "Asymmetric effects of positive and negative money supply shocks", Quarterly Journal of Economics, Vol. 107 No.4, pp.1261-82.
Diaz-Alejandro, C.F. (1963), "A note on the impact of devaluation and redistributive effects", Journal of Political Economy, Vol. 71 pp.577-80.
Engle, R.R. (1982), "A general approach to lagrange multiplier model diagnostics", Journal of Econometrics, Vol. 20 pp.83-104.
Falvin, M.A. (1981), "The adjustment of consumption to changing expectations about future income", Journal of Political Economy, Vol. 89 pp.974-1009.
Hall, R.E. (1978), "The stochastic implication of the life-cycle permanent income hypothesis: theory and evidence", Journal of Political Economy, Vol. 96 pp.971-87.
Hall, R.E., Mishkin, F.S. (1982), "The sensitivity of consumption to transitory income: estimates from panel data on households", Econometrica, Vol. 50 pp.461-81.
Heller, W.P., Starr, R.M. (1979), "Capital market imperfection, the consumption function, and the effect of fiscal policy", Quarterly Journal of Economics, Vol. 113 pp.455-63.
Hussein, K.A., DeMello, L.R. Jr (1999), "International capital mobility in developing countries: theory and evidence", Journal of International Money and Finance, Vol. 18 pp.367-81.
Kandil, M., Mirzaie, I.A. (2003), "The effects of dollar appreciation on sectoral labor market and adjustments: theory and evidence", The Quarterly Review of Economics and Finance, Vol. 43 pp.89-117.
Kandil, M., Mirzaie, I.A. (2006), "Consumption, and macroeconomic policies: theory and evidence from developing countries", Journal of International Trade and Economic Development, Vol. 15 No.4, pp.469-91.
Kandil, M., Mirzaie, I.A. (2007), "Consumption, credit, and macroeconomic policies: theory and evidence from the United States", unpublished manuscript, .
Kaynak, E., Kucukemiroglu, O., Ozmen, A. (1995), "Correlates of credit card acceptance and usage in an advanced developing middle eastern country", Journal of Services Marketing, Vol. 9 No.4, pp.52-63.
Krugman, P.R., Taylor, L. (1978), "Contractionary effects of devaluation", Journal of International Economics, Vol. 8 pp.445-56.
Kwiatkowski, D., Phillips, P.C.B., Schmidt, P., Shin, Y. (1992), "Testing the null hypothesis of stationarity against the alternative of a unit root: how sure are we that economic time series have a unit root?", Journal of Econometrics, Vol. 54 pp.159-78.
Newey, W., West, K. (1994), "Automatic lag selection in covariance matrix estimation", Review of Economic Studies, Vol. 61 pp.631-53.
Pagan, A. (1984), "Econometric issues in the analysis of regressions with generated regressor", International Economic Review, Vol. 25 pp.221-47.
Pagan, A. (1986), "Two stage and related estimators and their applications", Review of Economic Studies, Vol. 53 pp.517-38.
Reinhart, C.M., Vegh, C.A. (1995), "Nominal interest rates, consumption booms, and lack of credibility: a quantitative examination", Journal of Development Economics, Vol. 46 pp.357-78.
Sarno, L., Taylor, M.P. (1998), "Real interest rates, liquidity constraints, and financial deregulation: private consumption behavior in the UK", Journal of Macoreconomics, Vol. 20 pp.221-42.
Yin, Z., Hua Wan, G. (2002), "Household consumption and monetary policy in China", China Economic Review, Vol. 13 pp.27-52.
Further Reading
Benoit, C., Keita, S., Samoson, L. (1999), "Liquidity constraints and business cycles in developing countries", Review of Economic Dynamics, Vol. 2 pp.370-402.
Canner, G.B., Fergus, J.T. (1987), "The economic effects of proposed ceilings on credit card interest rates", Federal Reserve Bulletin, Vol. 73 pp.1-13.
Cashin, P., McDermott, C.J. (1998), "Testing the consumption-CAPM in developing equity markets", International Journal of Finance & Economics, Vol. 3 pp.127-41.
Dejuan, J.P., Seater, J.J. (1999), "The permanent income hypothesis: evidence from the consumer expenditure survey", Journal of Monetary Economics, Vol. 43 pp.351-76.
Friedman, M. (1957), A Theory of the Consumption Function, Princeton University Press for National Bureau of Economic Research, Princeton, NJ, .
Fuhrer, J.C. (2000), "Habit formation in consumption and its implications for monetary policy models", American Economic Review, June, .
Girardin, E., Sarno, L., Taylor, M.P. (2000), "Private consumption behavior, liquidity constraints and financial deregulation in France: a non-linear analysis", Empirical Economics, Vol. 25 pp.351-68.
Hahm, J-H. (1998), "Consumption adjustment to real interest rates: intertemporal substitution revisited", Journal of Economic Dynamics and Control, Vol. 22 No.2, pp.293-320.
Kandil, M. (2001), "Asymmetry in the effects of US government spending shocks: evidence and implications", The Quarterly Review of Economics and Finance, Vol. 41 pp.137-65.
Krugman, P.R., Baldwin, R.E. (1987), "The persistence of the US trade deficit", Brookings Papers on Economic Activity, No.1, pp.1-43.
Leong, K., McAleer, M. (1999), "Testing the life-cycle permanent income hypothesis using intra-year data for Sweden", Mathematics and Computers in Simulation, Vol. 48 pp.551-60.
Mathieson, D.J., Rojas-Suarez, L. (1990), "Financial market integration and exchange rate policy", IMF Working Paper WP/90/02, Washington, DC, .
Park, S., Rodrigues, A.P. (2000), "Is aggregate consumer borrowing consistent with the permanent income hypothesis?", The Manchester School, Vol. 68 No.3, pp.301-20.
Rodrik, D. (2000), "What drives public employment in developing countries?", Review of Development Economics, Vol. 4 No.3, pp.229-43.
Scott, A. (1996), "Consumption, ‘credit crunches’ and financial deregulation", Centre for Economic Policy Research, London, CEPR Discussion Paper, No. 1389, .
Zeldes, S.P. (1989), "Consumption and liquidity constraints: an empirical investigation", Journal of Political Economy, Vol. 97 pp.305-46.
Corresponding author
Magda Kandil can be contacted at: mkandil@imf.org
Appendix. Data sources
The sample period for investigation varies based on data availability as follows: Algeria (1963-2003), Egypt (1963-2002), Iran (1963-2002), Jordan (1966-2000), Libya (1963-2002), Oman (1967-2002), Pakistan (1963-2002), Syria (1964-2000), and Tunisia (1963-2003).
Variables used in investigation are as follows:
- Interest rate. Discount rate, International Financial Statistics (IFS), 60.ZF.
- Private consumption. Household consumption expenditure, IFS, 96F, … , ZF.
- Broad money. World Economic Outlook (WEO), WFMB.
- General government expenditure and net lending. WEO, WGCENL.
- Exchange rate. National currency per US$ WEO, WENDA.
- Real exchange rate. Nominal exchange rate multiplied by the US consumer price index (CPI) and divided by the developing countries CPI. The inverse measures the real value of national currency in terms of dollar. An increase indicates appreciation.
- GDP deflator. WEO, WNGDP.
- Nominal GDP. Gross domestic product current prices, WNGDP.
- Consumer price index. WEO, WPCPI.
All nominal variables have been deflated by the GDP deflator to measure real terms.
All country variables are from the IMF, IFS, or WEO, except for US CPI, which is taken from the Federal Reserve Bank of St. Louis.
See Tables AI-AIII overleaf.
