Printer Friendly

Oil price movements and GDP growth in GCC countries: is the relationship symmetric?


This paper examines whether oil price movements in one group of oil-exporting countries, namely those of the Gulf Cooperation Council, produce asymmetric effects on GDP growth as is commonly found with oil-importing countries. The paper employs a SUR model regressing annual growth rates against positive and negative changes in oil prices. Some evidence on the presence of asymmetries surfaces but is not overwhelming and is country specific.

JEL Classification: O13, O53

Keywords: Oil, Economic Growth, Asymmetric


No one doubts the importance of oil for the petroleum exporters of the Gulf Cooperation Council (GCC) countries. Nevertheless, questions do arise regarding magnitudes. How much does GDP growth increase following a rise in oil prices? Moreover, might oil price increases produce effects on growth differing in magnitude from oil price decreases? That is, is the relationship between the two asymmetric or are the effects of oil price declines simply the reverse of oil price increases? To our knowledge, there is no paper which investigates whether the relationship between oil price changes and GDP growth is symmetric or not for oil exporting countries.

Such an investigation is important for three reasons. For one, most studies examining the effects of oil price changes in oil-exporting countries implicitly assume a symmetric relationship. But if the relationship is asymmetric, then many of the inferences drawn from these studies could be skewed and so would need to be re-evaluated. Better understanding whether an asymmetry exists can also help guide policy. For example, if oil price movements have symmetric effects on GDP growth, then an x% decrease in oil prices followed by an x% increase has the same net effect on GDP growth as a constant oil price over the two periods. Thus, undergoing a mean preserving spread in the variance of oil price changes becomes less of a concern. But if effects are asymmetric, then changes in the variance become more of a concern, especially if negative changes in oil prices produce larger effects on the growth rate. To the extent that macroeconomic policy in these countries attempts to lower economic volatility due to oil price swings, then an asymmetric relationship would require that policy responses also be asymmetric.

Finally, several studies stemming from Hamilton (1983) have examined whether changes in oil prices produce asymmetric effects in oil-importing countries. (1) Some argue that changes in oil prices produce asymmetric effects on gasoline prices and this could be one channel as to why oil price movements produce asymmetric effects on output. (2) Others claim that monetary policy responds differently to oil price increases than to oil price decreases and this is the reason for the asymmetry. (3) Still others argue that oil price increases create sectoral imbalances and so cause greater adjustment costs than do oil price decreases. (4) A last group claims that oil price increases create greater uncertainty and this uncertainty worsens macroeconomic outcomes (5) But regardless of the specific reason, most (albeit not all) researchers examining oil-importing countries have found an asymmetric relationship between oil prices and GDP growth. A finding of asymmetry for oil exporting countries would suggest that these countries face similar (although opposite) dynamics as do oil-importers. On the other hand, a finding of symmetry would suggest that the relationship between oil prices and GDP growth in oil-exporting countries is not the mirror opposite of that between oil prices and GDP growth in oil-importing ones. Instead, oil exporting economies would face different qualitative dynamics following oil price shocks.

This study examines this issue for three countries in the Middle East, namely three of the six countries of the GCC: Kuwait, Saudi Arabia, and the United Arab Emirates (UAE) for the period 1964-2003. (6) Table 1 provides some summary statistics of oil production in these three countries. We focus attention on these countries because their similar histories and geographic proximity allow for a more homogeneous sample and so one that does not require as many controls in the empirical specification. (7) Moreover, these countries participate in a common economic union and so a specific analysis of this union seems warranted.

The remainder of the paper is organized as follows. Section 2 briefly discusses why asymmetries could be present in oil-exporting countries. Section 3 presents the empirical methodology and section 4 presents the results. Section 5 discusses implications and concludes the paper.


A priori, reasons exist not only for an asymmetry but for it to go in either direction. As in oil-importing countries, sectoral imbalances could arise from unfavorable oil price movements (decreases in the case of oil-exporting countries) that could cause costly reallocations of resources. One potential difference, however, is that the GCC economies, on the whole, are not as diversified as many oil-importing ones and so sectoral reallocations are less possible. The presence of borrowing constraints could also exacerbate effects of oil price decreases more so than oil price increases and so produce a negative asymmetry. For example, agents could both consume and save additional income stemming from oil price increases but, if savings levels are insufficient, only cut consumption following oil price decreases. If so, then the change in aggregate demand is greater following oil price decreases than after oil price increases.

However, a positive asymmetry could also arise due to movements in aggregate demand. Levels of government expenditure in these countries are largely determined by the amount of oil revenue. If governments find it politically easier to raise spending after increases in revenue than to reduce spending following decreases in revenue, then an asymmetry could go in the other direction in which positive oil price changes produce larger effects.

Of course, all of these effects could be small in magnitude or, if large, could offset one another. In either case, oil price movements would not have asymmetric affects on output and so, unlike for oil-importing countries, not skew effects on growth in one direction versus the other.


An unbalanced panel dataset is considered for these three GCC countries over the period 1964-2003. Table 2 presents oil price changes during this period. An advantage of using this time period is that positive (17) and negative (23) changes are both prevalent. There were 21 changes exceeding 10% in magnitude and 19 below this threshold. Moreover, 13 of these large changes were positive and 8 were negative. Thus, the sample is not overly skewed towards one type of change versus another. The dependent variable is the annual growth rate of real GDP, [y.sub.ti], where t denotes the year and i denotes the country.

A fixed effects model is employed where a separate intercept is used for each country. Thus, all time-invariant factors for growth in a country are implicitly captured. Later tests will also be conducted to determine if other coefficients in the specification are identical across countries and to determine if serial correlation is present in the residuals. Finally, the models are estimated-using a seemingly unrelated regression (SUR) technique so as to account for correlated disturbances across residuals.

The baseline specification comes from Mork (1989) who examines the effect of oil price percentage changes ([p.sub.t]) in oil-importing countries. Mork separates oil price movements into positive ([p.sub.t.sup.+]) and negative ([p.sub.t.sup.-]) percentage changes (where [p.sub.t.sup.+] = [p.sub.t] when [p.sub.t] > 0 and [p.sub.t.sup.+] = 0 when [p.sub.t] [less than or equal to] 0 and where [p.sub.t.sup.-] = [p.sub.t] when [p.sub.t] < 0 and [p.sub.t.sup.-] = 0 when [p.sub.t] [greater than or equal to] 0). The model is:

[y.sub.ti] = [a.sub.i] + [bi][p.sub.t-1.sup.+] + [c.sub.i][p.sub.t-1.sup.-] + [d.sub.i][y.sub.t-1,i] + [e.sub.ti] (1)

where e denotes the unobservable and where y denotes the growth rate of GDP. We allow for a different set of coefficients for each country and will later test for common coefficients. The null hypothesis of symmetry is that b = c. The lag of the growth rate is also included to capture persistent effects on growth from the previous year. We consider lagged oil price changes to diminish problems of reverse causality as economic performance could influence supplies of oil and so world prices. Given that (1) includes the lagged growth rate, [y.sub.t-1], the coefficients b and c solely capture direct effects of lagged changes in oil prices on current growth and not indirect effects through the past growth rate. In other specifications, we remove lagged growth from (1) by setting d to zero. Although this exclusion does not allow us to control for persistent influences on growth, it does allow for the coefficients b and c to capture the total effect of lagged oil price changes on growth.

The model in (1) is simple in that many other variables are excluded (although, as stated above, the fixed effect captures all time invariant factors), more so than in Mork (1989). Their exclusion stems from two reasons. For one, other factors such as monetary policy are less important for these countries than for the U.S. Second, they are potentially endogenous. For example, given that oil prices affect oil revenue which determines government expenditure, including government spending in the model would mask some of the effect that oil prices have on GDP growth rates. Given that we are interested in capturing the total effect from oil price movements to determine if any asymmetry is present, we find this disadvantage of including such variables to outweigh the advantages. However, including the past growth rate in (1) helps implicitly capture persistent effects on the growth rate. (8)

An implicit assumption in (1) is that oil prices have linear effects on output growth. To account for possible nonlinear effects, the lags of two dummy variables will be added to (1) which divides oil price movements into "small" and "large" ones. For the first dummy, let [j.sub.t] = 1 when [p.sub.t] > 10% and zero otherwise. And for the second, let [k.sub.t] = 1 when [p.sub.t] < -10% and zero otherwise. Thus, the first dummy takes the value one in the presence of a large positive change to oil prices and the second dummy equals one when a large negative oil price movement occurs. Inclusion of these dummies allows for the possibility of small changes to have essentially zero effect on output growth whereas large changes can have sizable influences should the coefficients on these dummies be large in magnitude. The null hypothesis of symmetry is that the coefficients for the lags of these dummy variables are oppositely signed but are equal in magnitude. Admittedly, ten percent is an arbitrary threshold. Given the distribution of oil price changes in table 2, a 10%+ change of magnitude happens roughly 50% of the time. Thus, such changes are not mere outliers but do not dominate the sample either.

The final specification follows Hamilton (1996) who allows for "unusual" oil price movements to have distinct effects on growth. Analogous to his variable (which accounts for unusual oil price increases), we define: [ep.sub.t] = min{0, [P.sub.t] - min ([P.sub.t-1], [P.sub.t-2], [P.sub.t- a])} where [P.sub.t] denotes the oil price level in year t and ep denotes an unusual oil price decrease. ep equals zero unless the current year's average oil price is lower than the average oil price in any of the most recent three years in which case ep equals the difference between the current average price and the previous minimum average of the last three years. The assumption is that such a low price lies outside the recent norm. The variable ep is then added to (1) and a significant coefficient would indicate that these unusual price decreases have different effects on growth than do other downward movements.

GDP growth data comes from the World Bank's World Development Indicators and is adjusted for inflation. Oil price data comes from OPEC and is also adjusted for inflation using consumer price indices of modified Geneva I countries and the USA. (9) The price data comes from an index where 1973 = 100. The change for year t is calculated as the percentage difference in the year t+1 January average and the year t January average. Until 1981, the price of Arabian light is used to capture the price of oil and then the OPEC spot reference basket price is used afterwards.


Before estimating a fixed effects model, we consider each of the three countries separately. The results are presented in Table 3. Evidence of serial correlation within the residual is weak except for Saudi Arabia (without including the lagged growth rate) where we correct for serial correlation using an AR(2) structure for the residual. The other specifications were not adjusted for serial correlation.

To put the coefficients on the oil price variables into perspective, consider a hypothetical coefficient of 0.1 for [p.sup.+] which is not unrealistic given the individual country coefficients in the table. Such a magnitude indicates that a 10% increase in the price of oil raises GDP growth by one percentage point. Similarly, a coefficient of 0.1 for p would indicate that a 10% decrease in the price of oil lowers growth by one percentage point. Given the obvious importance of oil for these economies, we find these coefficient estimates reasonable.

For Kuwait, neither positive nor negative lagged changes in oil prices are strongly correlated with growth. The null hypothesis of symmetry is not rejected. For Saudi Arabia, the null hypothesis of symmetry is rejected for the simpler specification but not once we include lagged growth. The coefficient b is significantly positive indicating that positive oil changes increase growth. The negative values for c indicate that even negative changes are still positively associated with growth although the higher standard errors considerably temper this inference. The largest values of b and c occur for the UAE where negative oil price changes greatly lower growth. The null hypothesis of symmetry is rejected under the simpler specification but not when including the lagged growth rate. (10)

For the specifications in Table 3, we examine the possibility of using a panel data set so as to estimate the equations jointly but where b, c, and (when relevant) d are identical across countries. However, F-statistics from specification tests reject the null of the restricted model (common coefficients) versus the alternative of differing coefficients across countries. Hence, we continue to employ separate coefficients. But we estimate (1) by SUR since common shocks to the region could cause the respective residuals to be correlated. Estimating (1) by OLS presents similar findings.

The first two columns of Table 4 repeat the above regressions except using a SUR model. Findings coincide with those above. The next two columns add the lags of the j and k dummies which take the value one for large positive or negative changes in oil prices. Evidence of nonlinearities exists for both Saudi Arabia and the UAE although only for the latter is evidence of asymmetries present. Positive oil changes now have stronger effects than negative ones and contrast earlier findings. Hence, small negative changes in oil prices seem to matter more for growth than do small positive changes, at least for the UAE and perhaps Saudi Arabia. However, no strong evidence arises of nonlinear asymmetries stemming from large changes to oil prices. (11)

The last two columns of Table 4 add the ep variable but its coefficient is never significant in any of the three countries. There is no evidence that unusual downward changes in oil prices have additional adverse effects on growth.


Much past research has found an asymmetric effect between oil price movements and GDP growth in oil-importing countries. Our findings for oil-exporting countries are mixed and we find no general evidence of asymmetries. To the extent that asymmetries are not present, this then suggests that the effects of oil price movements on GDP growth rates are not only opposite but qualitatively differ between oil exporters and importers. One possible reason is that oil production and sales for these GCC countries comprise larger relative shares of their economies than does oil consumption and imports for most, if not all, oil-importing countries. If the reason for asymmetries in oil-importing countries is due to sectoral adjustments, then such effects might not exist in countries where one sector dominates the economy and so provides one possible explanation for differences between the two sets of countries.

An implication is that these oil-exporters might, consequently, be less concerned about oil price stability than are oil-importing countries. For example, under symmetry a large increase followed by a decrease in oil price of similar magnitude will have similar net effects on growth as a constant oil price. Although there could be other reasons why oil-exporters would prefer more stable prices, this study finds no strong evidence that these GCC countries see GDP growth rates fall on average due to oil price instability.

Moreover, the results also suggest that policy makers to the extent that they want to stabilize GDP growth rates need not be more aggressive following oil price decreases than after oil price increases. Monetary and fiscal policies can be similar in magnitude (although, of course, opposite in direction).

Finally, to the extent that borrowing constraints would exacerbate effects from negative income shocks, then this study does not find strong evidence that these constraints are binding for these GCC countries.

However, the UAE could be an important exception in that negative changes in oil prices have bigger effects than do positive changes. To this extent, the UAE would prefer to see more stable oil prices and so could lead to different preferences over policy within GCC and, more generally, OPEC countries.


Aizenman, J. and Marion N. (1993), "Macroeconomic Uncertainty and Private Investment," Economics Letters 41, 207-210.

Aizenman, J. and Marion, N. (1999), "Volatility and Investment: evidence from developing Countries," Economica 66, 157-179.

AlGudhea, S. (2003), "Essays on the Behavior of the Oil Market and OPEC," Dissertation, Southern Illinois University at Carbondale, Department of Economics.

Bacon, R. (1991), "Rockets and Feathers: the Asymmetric Speed of Adjustment of UK Retail Gasoline prices to Cost Changes," Energy Economics 13, 211-18.

Ball, L. and Mankiw, G. (1994), "Asymmetric Price Adjustment and Economic Fluctuations," Economic Journal 104, 247-261.

Bernanke, B. (1983), "Irreversibility, Uncertainty, and Cyclical Investment," Quarterly Journal of Economics 98, 85-106

Bernanke, B., Gertler, M. and Watson, M. (1997), "Systematric Monetary Policy and the Effects of Oil Price Shocks," Brookings Papers on Economic Activity 1, 91-157.

Bohaha, A. (1988), "An Analysis of the Effects of Oil Price Shocks on the Colorado Economy Using Vector Autoregression," Review of Regional Studies 8, 23-30.

Borenstein, S., Cameron, A. and Gilbert, R. (1997), "Do Gasoline Prices Respond Asymmetrically to Crude Oil Price Changes?" Quarterly Journal of Economics, 112, 305-339.

Davis, S. and Haltiwanger, J. (2001), "Sectoral Job Creation and Destruction Response to Oil Price Changes," Journal of Monetary Economics 48, 465-512.

Davis, S., Loungani, P. and Mahidhara, R. (1997), "Regional Labor Fluctuations: Oil Shocks, Military Spending and Other Driving Forces," International Finance Discussion Paper number 578, Board of Governors, Federal Reserve System.

Ferderer, J. (1996), "Oil Price Volatility and the Macroeconomy: A Solution to the Asymmetry Puzzle," Journal of Macroeconomics 18, 1-26.

Gardner, T. and Joutz, F. (1996), "Economic Growth, Energy Prices and Technological Innovation," Southern Economic Journal 62, 653-666.

Gately, D. (1993), "The Imperfect Price-Reversibility of World Oil Demand," Energy Journal 14, 163-182.

Gately, D. and Huntington, H. (2002), "The Asymmetric Effects of Changes in Price and Income on Energy and Oil Demand," Energy Journal 23, 19-55.

Griffin, J. and Schulman, C. (2004), "Price Asymmetry: A Proxy for Energy Saving Technical Change?" Working Paper #405, The Bush School of Government & Public Service, Texas A&M University.

Hamilton, J. (1983), "Oil and the Macroeconomy Since World War II," Journal of Political Economy 91, 228-248.

Hamilton, J. (1988), "A Neoclassical Model of Unemployment and the Business Cycle," Journal of Political Economy 96, 593-617.

Hamilton, J. (1996), "This is What Happened to the Oil Price-Macroeconomy Relationship," Journal of Monetary Economics 38, 215-220.

Hamilton, J. (2000), "What is an Oil Shock?" Journal of Econometrics 113, 363-398.

Hamilton, J. and Herrera, A. (2001), "Oil Shocks and Aggregate Macroeconomic Behavior: The Role of Monetary Policy," Discussion Paper 2001-10, Department of Economics, University of California, San Diego.

Hooker, M. (2002), "Are Oil Shocks Inflationary? Asymmetric and Nonlinear Specifications versus Changes in Regime," Journal of Money, Credit & Banking 34, 540-561.

Huntington, H. (1998), "Crude Oil Prices and U.S. Economic Performance: Where Does the Asymmetry Reside?" Energy Journal 19, 107-132.

Keane, M. and Prasad, E. (1996), "The Employment and Wage Effects of Oil Price Changes: A Sectoral Analysis," Review of Economics and Statistics 78, 389-400.

Kirchgassner, G. and Kubler, K. (1992), "Symmetric or Asymmetric Price Adjustments in the Oil Market: An Empirical Analysis of the Relations between International and Domestic Prices in the Federal Republic of Germany, 1972-89," Energy Economics 14, 171-185.

Lee, K., Ni, S. and Ratti, R. (1995), "Oil Shocks and the Macroeconomy: the Role of Price Variability," Energy Journal 16, 39-56.

Lilien, D. (1982), "Sectoral Shifts and Cyclical Unemployment," Journal of Political Economy 90, 777-793.

Loungani, P. (1986), "Oil Price Shocks and the Dispersion Hypothesis," Review of Economics and Statistics 68, 536-539.

Mork, K. (1989), "Oil and the Macroeconomy When Prices Go Up and Down: An Extension of Hamilton's Results," Journal of Political Economy, 97, 740-744.

Mory, J. (1993), "Oil Prices and Economic Activity: Is the Relationship Symmetric?" Energy Journal 14, 151-161.

Mork, K., Olsen, O. and Mysen, T. (1994), "Macroeconomic Responses to Oil Price Increases and Decreases in Seven OECD Countries," Energy Journal 15, 19-35.

Pindyck, R. (1991), "Irreversibility, Uncertainty, and Investment," Journal of Economic Literature 29, 1110-1148.

Sadorsky, P. (1999), "Oil Price Shocks and Stock Market Activity," Energy Economics 21, 449-469.

Tatom, J. (1987), "The Macroeconomic Effects of the Recent Fall in Oil Prices," Federal Reserve Bank of St. Louis Review 69, 34-45.

Tatom, J. (1988), "Are the Macroeconomic Effects of Oil Price Changes Symmetric?" Carnegie-Rochester Conference Series on Public Policy 28, 325-368.

Tatom, J. (1993), "Are There Useful Lessons from the 1990-1991 Oil Price Shock?" Energy Journal 14, 129-150.


(1.) See Tatom (1987) Mork (1989), Gately (1993), Mory (1993), Mork et al. (1994), Lee et al. (1995), Federer (1996), Gardner and Joutz (1996), Sadorsky (1999), Davis and Haltiwanger (2001), Griffin and Schulman (2004), Gately and Huntington (2002) for contributions to this literature.

(2.) See Bacon (1991), Kirchgassner and Kubler (1992), Borensteirn et al. (1997), Huntington (1998), and Algudhea (2003) for findings for and against this hypothesis.

(3.) See Tatom (1988, 1993) Bernanke et al. (1997), Hamilton and Herrera (2001), and Hooker (2002).

(4.) See Lilien (1982), Loungani (1986), Bohaha (1988), Hamilton (1988), Ball and Mankiw (1994), Keane and Prasad (1996), and Davis et al. (1997) for further details.

(5.) See Bernanke (1983), Pindyck (1991), Aizeman and Marion (1993, 1999), and Hamilton (2000).

(6.) Six countries comprise the GCC. However, data for Qatar is not available and data for Bahrain is only available after 1980. Data for Oman is available but Oman produces less oil both absolutely and relative to GDP than do the remaining three countries. Oman is also not a member of OPEC. When Oman was considered, no asymmetric effects were found. An additional reason to focus on Kuwait, Saudi Arabia, and the UAE is that these comprise three of the seven top oil exporters in the world.

(7.) As counterexamples, OPEC contains less similar countries such as Venezuela and Nigeria. Moreover, Russia, another large oil-exporter, underwent tremendous transition during the 1990's and so including it in the sample does not seem appropriate.

(8.) Including the lag of the dummy variable, d73, in (1) which equals one in 1973 and zero otherwise so as to capture the extreme increase in oil prices during this year, 208%, did not change the results. In addition, growth data is missing for Kuwait for 1990-1992. Therefore, we do not make any adjustment for the Iraqi invasion of Kuwait during this time.

(9.) Using oil price data that is also adjusted for exchange rate changes using the Geneva I and USA basket of currencies versus a basket of currencies using merchandise imports from OPEC countries as weights did not change the findings.

(10.) We also examined estimations where oil price changes were not split between positive and negative ones but where GDP growth was regressed against lagged oil price changes with or without the lagged growth rate. Lagged oil price changes were significantly correlated with GDP growth in Saudi Arabia and the UAE but not so for Kuwait, suggesting that oil price changes have mattered more for the former countries, at least in the short run.

(11.) We also examined the presence of nonlinearities by including the squares of p+ and p- but continued not to find strong evidence of asymmetry.


Saudi Arabia Monetary Agency, Kingdom of Saudi Arabia


Southern Illinois University, Carbondale
Table 1
Average Daily Oil Production

Year Kuwait Saudi UAE countries *

 Oil production as a percent of
 OPEC production

1960-70 15.64 15.82 1.92 --
1971-80 8.53 22.23 5.71 --
1981-90 6.50 28.31 7.51 --
1991-00 6.56 31.12 8.61 --
2001-04 7.49 29.79 8.10 --

 Oil production as a percent of total
 world oil production

1960-70 7.34 7.45 0.92 0.96
1971-80 4.35 11.35 2.91 1.10
1981-90 2.24 9.78 2.60 0.95
1991-00 2.73 12.95 3.58 1.12
2001-04 3.06 12.16 3.30 1.46

Source: Energy Information Administration (

* Bahrain and Oman are not members of OPEC.

Table 2

Changes in Real Oil Prices

Year % Change

1960 -5.45%
1961 -3.32%
1962 -3.62%
1963 -3.11%
1964 -3.75%
1965 -3.50%
1966 -1.68%
1967 -3.61%
1968 -4.28%
1969 -5.17%
1970 15.43%
1971 7.24%
1972 22.67%
1973 207.74%
1974 -10.90%
1975 -2.26%
1976 -2.07%
1977 -4.67%
1978 23.80%
1979 46.96%
1980 2.06%
1981 -8.52%
1982 -15.94%
1983 -8.52%
1984 -9.19%
1985 -51.46%
1986 26.77%
1987 -22.48%
1988 15.63%
1989 21.58%
1990 -19.92%
1991 -4.75%
1992 -14.24%
1993 -7.59%
1994 5.25%
1995 17.17%
1996 -9.92%
1997 -35.41%
1998 40.23%
1999 54.10%
2000 -18.12%
2001 3.42%
2002 12.80%
2003 25.47%
2004 37.24%

Table 3
Separate Least Squares Regressions


Variable Coeff. Kuwait Kuwait S. Arab. (a)

Intercept a 3.817 * 2.944 2.614
 (2.182) (2.078) (4.009)
lag of [p.sup.+] b -0.069 -0.062 0.071 **
 (0.050) (0.045) (0.027)
lag of [p.sup.-] c -0.017 0.024 -0.118
 (0.113) (0.151) (0.081)
[y.sub.t-l,i] d 0.093

 P-Values from following tests

Serial Correlation (b) 0.997 0.42 0.27 (c)
Wald test of b = c 0.77 0.61 0.04
# of observations 38 36 33


Variable S. Arab. UAE UAE

Intercept 1.249 8.576 5.630 **
 (1.730) (1.902) (2.322)
lag of [p.sup.+] 0.061 * 0.051 0.231 **
 (0.035) (0.036) (0.094)
lag of [p.sup.-] -0.025 0.492 *** 0.374 ***
 (0.109) (0.121) (0.124)
[y.sub.t-l,I] 0.556 *** 0.191
 (0.148) (0.135)

 P-Values from following tests

Serial Correlation (b) 0.33 0.54 0.61
Wald test of b = c 0.48 0.001 0.44
# of observations 34 29 28

Standard Errors in parentheses

(a) After including an AR(2) error specification

(b) Breusch-Godfrey test for no serial correlation up to 2 lags

(c] Without AR(2) error specification, p-value is 0.0006

***, **, * denotes significance at 1%, 5%, and 10% levels, respectively

Table 4
SUR Results

 (1) (2) (3)


intercept 3.830 * 3.428 * 2.556
 (2.091) (1.931) (2.423)
lag of [p.sup.+] -0.067 -0.064 -0.098 *
 (0.048) (0.042) (0.055)
lag of [p.sup.-] -0.020 0.046 0.026
 (0.158) (0.142) (0.257)
[y.sub.t-1] 0.026
lag of j 4.466
lag of k 3.387
lag of ep


Wald tests
b = c 0.79 0.48 0.64
r = s 0.33
# of obs. 38 36 38

 S. Arabia (a)

intercept 3.384 1.625 2.168
 (3.436) (1.597) (3.524)
lag of [p.sup.+] 0.073 *** 0.060 * 0.056 **
 (0.024) (0.031) (0.028)
lag of [p.sup.-] -0.089 -0.011 -0.161
 (0.076) (0.101) (0.138)
[y.sub.t-1] 0.544 **
lag of j 2.615
lag of k -1.528
lag of ep


Wald tests
b = c 0.06 0.52 0.11
r = s 0.78
# of obs. 33 34 33


intercept 9.254 *** 5.768 *** 6.181 ***
 (1.777) (2.061) (1.936)
lag of [p.sup.+] 0.056 0.220 *** 0.005
 (0.034) (0.092) (0.036)
lag of [p.sup.-] 0.508 *** 0.408 *** 0.517 ***
 (0.113) (0.112) (0.162)
[y.sub.t-1] 0.253 **
lag of j 8.538 **
lag of k 4.086
lag of ep


Wald tests
b = c 0.0004 0.25 0.002
r = s 0.02
# of obs. 29 28 29

 (4) (5) (6)


intercept 1.630 2.900 2.723
 (2.190) (2.239) (1.957)
lag of [p.sup.+] -0.099 ** -0.063 -0.065
 (0.047) (0.047) (0.040)
lag of [p.sup.-] -0.222 -0.025 0.054
 (0.233) (0.156) (0.138)
[y.sub.t-1] -0.002 -0.069
 (0.148) (0.152)
lag of j 5.629
lag of k -7.005
lag of ep -0.219 -0.248
 (0.183) (0.169)


Wald tests
b = c 0.60 0.83 0.44
r = s 0.85
# of obs. 36 37 36

 S. Arabia (a)

intercept 0.166 4.516 2.476
 (1.969) (3.148) (1.659)
lag of [p.sup.+] 0.038 0.072 *** 0.060 *
 (0.035) (0.023) (0.030)
lag of [p.sup.-] -0.182 -0.053 -0.002
 (0.170) (0.077) (0.099)
[y.sub.t-1] 0.594 *** 0.520 ***
 (0.134) (0.130)
lag of j 3.235
lag of k -3.722
lag of ep 0.209 ** 0.159
 (0.100) (0.116)


Wald tests
b = c 0.19 0.14 0.57
r = s 0.92
# of obs. 34 33 34


intercept 3.962 *** 10.043 *** 5.723 **
 (2.089) (1.847) (2.404)
lag of [p.sup.+] -0.003 0.054 0.219 ***
 (0.147) (0.033) (0.081)
lag of [p.sup.-] 0.400 ** 0.521 *** 0.408 ***
 (0.162) (0.111) (0.114)
[y.sub.t-1] 0.279 ** 0.227 *
 (0.116) (0.149)
lag of j 10.494 *
lag of k 2.019
lag of ep 0.155 -0.008
 (0.126) (0.149)


Wald tests
b = c 0.06 0.0002 0.25
r = s 0.06
# of obs. 28 29 28

Standard Errors in parentheses

(a) Regressions without lagged growth employ an AR(2) error
specification b and c denote the coefficients on [p.sup.+] and
[p.sup.-], respectively. r and s denote the coefficients on j and k,

***, **, * denotes significance at 1%, 5%, and 10% levels, respectively
COPYRIGHT 2007 Indian Journal of Economics and Business
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Al-Otaibi, Bader; Sylwester, Kevin
Publication:Indian Journal of Economics and Business
Geographic Code:70MID
Date:Dec 1, 2007
Previous Article:Does U.S. monetary policy react to asset prices? Implications of stock market bubbles, volatility and productivity.
Next Article:Successful replication of the article: buffer-stock money: interpreting short-run dynamics using long-run restrictions.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters