Abstract

Whether or not a current account deficit is sustainable has important implications for policy. If the current account deficit of a nation is sustainable, then it implies that the government should have no incentive to default on its international debt. In this article, we examine whether or not the current account deficits of the Economic and Monetary Community of Central African countries are sustainable. The econometric methodology adopted threshold cointegration test advanced by Enders and Siklos [7]. The findings indicate that: (1) the current account deficits are only strong sustainable for Chad and weakly sustainable for Central Africa Republic, Congo and Gabon; (2) the results from panel cointegration test showed that current account deficits are weakly sustainable in CEMAC area; (3) Cameroon and Equatorial Guinea are in violation of their international budget constraints and should therefore put in place policies to reduce their current account deficits in order to regain their external stability; and (4) For the other country members of CEMAC where sustainability was found to be weak, they should also implement policies to reinforce the sustainability of the current account deficits.

1. Introduction

Maintaining balanced current accounts is a tremendous challenge that faces many developing countries in general and African countries in particular. This is because, on one hand, most countries depend largely on imports of various types of goods for the consumption of both the rapidly growing populations and investment sectors. On the other, their export sector is still uncompetitive in the world markets. Consequently, the growing gap between rapidly expanding imports and shrinking or stagnant export sectors causes current account deficits to be unsustainable over times since current account deficits are a reflection of either the strength of domestic economies as they measure resource inflows into a country or an accumulation of debts that cannot be serviced in the future. A mixture of external and internal factors such as unsustainable fixed exchange rate regimes, real exchange rate appreciation, and continuous deterioration of terms of trade and current account imbalances, short-term foreign debts and weak financial system are believed to cause persistent current account deficits.

The concept of current account simply represents an analysis of whether the country is capable of fulfilling its long-run intertemporal budget constraint without triggering dramatic economic upheavals that can cause painful adjustments. Long episodes of unsustainable current fluctuations can doubtlessly initiate either sudden exchange rate crisis that is reminiscent of those observed in Chile in 1982, Mexico in 1994-95 and the Southeast Asia countries in 1997 or an accumulation of foreign debts that the country may not be able to service in the long run, eventually triggering a crisis. Furthermore, large and growing current account deficits usually cause negative net international investment position to grow larger, eventually generating financial burden such as interest rates and dividends to the extent of adversely affecting domestic consumption and business investment. Mann (2002) described current account sustainability as a situation in which external imbalances do not generate economic forces that can steer the economy away from its long-term path.

When countries run large current account deficits, there is always general nervousness on the side of financial world as to whether these countries could be able to sustain such deficits for long time. This raises the need for some rule or criterion for gauging whether a given current account deficit is sustainable or not. In this respect, Milesi-Ferretti and Razin ¹⁴ suggested that a current account deficit of say, 5 percent of gross domestic product (GDP), should constitute an appropriate alarm bell to policy makers. They acknowledged the need for such a rule especially when the deficit is financed with short-term debts or foreign exchange reserves and particularly when the deficit reflects the high consumption spending rather than investment although they questioned the seriousness with which such rule should be taken.

However, a closer look at historical episodes of economic crises revealed that not all large current-account deficits always lead to external crises. Experience shows that a number of countries that included Australia, Ireland, Israel, Malaysia and South Korea were able to sustain large current-account deficits for a number of years without necessarily triggering large-scale currency crisis. However, others such as Chile, Mexico and the Southeast Asian countries (especially Malaysia, the Philippines and Thailand) suffered severe crisis in 1997. As deficit/GDP ratio is not a reliable rule, Milessi-Ferretti and Razin ¹⁴ further suggested the country’s solvency for evaluating the unsustainability of large current account deficits. That is, does the country have the ability to generate sufficient trade surpluses to repay its accumulated debts in the future? Furthermore, they drew a distinction between solvency and sustainability by arguing that an economy is considered solvent if the expected present value of future trade surpluses equals the economy’s current accumulated debt. On the one hand, the notion of solvency should be satisfied when the country satisfies its intertemporal budget constraint, that is, as long as the discounted value of country’s foreign debt is non-zero in the infinite limit provided that the country should not increase its foreign debt faster than the real interest rate on the accruing debt. This criterion implies that a country could run very large current account deficits and be solvent provided that first, there are expected surpluses in the future and second, the country’s stock of foreign debt can increase without limits as long as the country does not increase it faster than the real interest rate. On the other, a sustainable current account is one that does not trigger feedbacks on domestic saving and investment or cause significant international portfolio reallocations that leads to a rise in interest rates. Thus, when a country runs a persistent large and growing current account deficit, it would inevitably generate changes in GDP growth and thus, in import spending, causing its present level to be unsustainable. However they expressed doubts about the appropriateness of solvency concept as a criterion for evaluating the country’s current account sustainability since it considers only the country’s ability rather than its willingness to pay. Although the present value of trade surpluses could be sufficient to repay the country external debts, theoretically, however, diverting output from domestic consumption to debt servicing can possibly involve high political risks. In addition, the concept relies on unrealistic assumptions that foreign investors are willing to lend to the country on current terms irrespective of the country’s ability to meet its debt obligation. In absence of external shocks and macroeconomic policy changes, current account is sustainable only as long as no external sector crisis occurs.

Roubini and Wachtel ²¹ who considered this theoretical current-account sustainability criterion not to be stringent enough since the intertemporal budget constraint only imposes very mild restrictions on the evolution of the country’s current account and foreign debt suggested the non-increasing foreign debt to GDP ratio as the practical sufficient condition for sustainability. However, they recognized the need to distinguish between current account deficits that arise from growth-inducing capital inflows and current account deficits that result in debt accumulation that cannot be sustained. Based on these considerations, a sustainable current account deficit is, therefore, one that can be maintained without any of these crises occurring.

When testing for cointegration between imports and exports, most of studies use linear cointegration tests of Engle and Granger ⁸, Johansen ¹³ and Pesaran et al. ¹⁷ all of which assume symmetric adjustment towards the long-run equilibrium. According to Balke and Fomby ², because of the presence of transaction costs and asymmetries in price transmission, there is no reason to expect a symmetric adjustment. They therefore initiated the idea of threshold cointegration where the adjustment towards the long-run equilibrium occurs only when the deviation from the equilibrium exceeds some threshold ²³. Following Balke and Fomby ², Enders and Granger ⁶ and Enders and Siklos ⁷ employed threshold Autoregressive (TAR) and Momentum threshold Autoregressive (M-TAR) models to develop a threshold cointegration test which allows for asymmetric adjustment towards the long run equilibrium.

The objective of this paper is to establish the relationship between imports and exports using the threshold cointegration test advanced by Enders and Siklos ⁷ to test the sustainability of the current account deficits in the six African countries of CEMAC area. For this purpose, we organize the rest of the paper as follow; Section II presents the conceptual framework; section III, the data and methodology used. Section IV presents and discusses the results and section V gives the concluding remarks.

2. The Conceptual Framework

Testing for the cointegration relationship between imports and exports as a way of checking the sustainability of current account deficits was first proposed by Hakkio and Rush ⁹ and Husted ¹¹.

They proposed a conceptual framework in which a representative individual of a small open economy faces the following budget constraint:

(1)

Where , and stand for current consumption, income and Investment respectively. is the current borrowing, is the initial debt size and is the world interest rate.

Solving for in equation (1) yields expression (2) where the trade balance and is the discounting factor:

(2)

To get a testable equation, Husted ¹¹ makes the following assumption where and is expenditure on imports:

(3)

From equation (3), solving for yields:

(4)

Husted ¹¹ assumes further that expenditure on imports and exports are non-stationary processes which can be written as:

(5)

(6)

Substituting equations (5) and (6) in equation (4) and rearranging gives:

(7)

By letting and

equation (7) can be written as:

(8)

Finally, equation (8) can be written as follows where and assuming that :

(9)

According to Hakkio and Rush ⁹ and Husted ¹¹, the current account deficits are sustainable if exports , and imports , are cointegrated. It has been argued however that for the current account deficits to be strongly sustainable, the sufficient condition should be that and in case, they are only weakly sustainable ^{10, 20, 22, 25}

3. Data and Methodology

We used annual data on exports and imports (in logarithm) for the six Economic and Monetary Community of Central African countries namely Cameroon, Chad, Central African Republic, Congo, Gabon and Equatorial Guinea covering the period 1960-2014. Data were collected from World Development Indicator 2016. Table 1 presents descriptive statistics of exports and imports (in logarithm) by country and Figure 1, their evolution on the entire period of study. As we see, mean (respectively, standard deviation, minimum and maximum) of exports and imports by country are close to each other. And the graph seems to indicate that imports and exports have a co-movement between them, giving an impression that there might be a cointegration relationship between them in the countries under study.

Threshold cointegration technique initiated by Enders and Granger ⁶ and Enders and Siklos ⁷ is presented hereafter, method which is employed in this study to test for cointegration between imports and exports in our Central African Countries.

Extending Engle & Granger’s ⁸ linear cointegration test, Enders and Granger ⁶ and Enders and Siklos ⁷ developed a threshold cointegration test where negative and positive deviations from the long-run equilibrium are not corrected in the same way, that is, in which the adjustment towards the long-run equilibrium is asymmetric ²³.

Let and be the logarithm of exports and imports respectively. Using TAR and M-TAR models, Enders and Siklos ⁷ propose the following steps to test for threshold cointegration.

In the first step, the following long-run equilibrium relationship is estimated:

(10)

In the next step, the following equation is estimated using Ordinary Least Squares (OLS):

(11)

whereis the residuals series from equation (10) and is the Heaviside indicator function such that:

(12)

(13)

Where is the threshold value to be estimated.

Equations (11) and (12) together form the threshold autoregressive model (TAR) and equations (11) and (13) form the momentum threshold autoregressive model (M-TAR). The threshold value is selected using Chan’s ³ method where the optimum value is such that the residuals sum of squares is at a minimum ²⁴. From equation (11), to test for threshold cointegration, Enders and Granger ⁶ and Enders and Siklos (2001) propose to test the following hypothesis of no threshold cointegration:

The test statistic used is known as statistic and the critical values are from Enders and Siklos ⁷.

4. Empirical Investigations

Before running the cointegration test, the variables must be tested for stationarity. For this purpose, the traditional Augmented Dickey-Fuller ⁴ test is used to test for the order of integration of exports and imports in CEMAC countries. Unit root test results are presented in Table 2 and suggest that exports and imports (in logs) for all the CEMAC area are nonstationary processes becoming stationary after one differentiation. Since imports and exports are both integrated of order 1 for all the countries under study, we can test for cointegration relationship between them.

Table 3 and Table 4 present the results of threshold cointegration test between imports and exports for the CEMAC countries, using TAR and momentum TAR models. The optimal threshold value λ minimizing the residuals sums of squares was estimated using Chan’s ³ method. For the TAR model for instance the estimated threshold value is for Cameroon, for Central Africa Republic, for Chad, for Congo, for Equatorial Guinea and for Gabon. For the M-TAR model, the estimated threshold value for each country is reported in Table 4. The results in Table 3 and Table 4 indicate also that Ljung-Box test fails to reject the null hypothesis of no serial correlation at 5 % level of significance.

Using Akaike Information Criterion (AIC), the number of lags k to include in the TAR and M-TAR models was also selected. For the M-TAR model for instance, out of a maximum of 7 lags, AIC selects a lag of 1 for Cameroon and Chad and a lag of 0 for the rest of countries. It should be noted that for the TAR model, AIC selects also the same lags.

We took into account the estimated threshold value and optimal lag length selected to test for threshold cointegration between imports and exports.

Threshold cointegration tests results based on the TAR model are reported in Table 3. They indicate that the test statistic rejects the null hypothesis of no threshold cointegration at 5% level of significance for Central Africa Republic, Chad, Congo and Gabon and at 10% level of significance for Cameroon and Equatorial Guinea. Thus, the estimated TAR model for Chad (for example) can be written as follows with standard deviation in parentheses:

where

Based on the Momentum TAR model, the results of threshold cointegration are reported in Table 4. They show that the null hypothesis of no threshold cointegration can be rejected at 5% level for all countries except Equatorial Guinea. Thus, the estimated M-TAR model for Chad (for example) can be written as follows:

where

Together, Table 3 and Table 4 also indicate that the null hypothesis of no threshold cointegration is rejected for Central Africa Republic, Chad, Congo and Equatorial Guinea at 5% level for both TAR and M-TAR models. However, for Cameroon and Equatorial Guinea, the null hypothesis of no cointegration could not be rejected at 5% level for both TAR and M-TAR model.

The results suggest therefore that imports and exports are cointegrated with asymmetric adjustment for Central Africa Republic, Chad, Congo and Gabon. This would imply that the current account deficits in those countries are sustainable. However, we need to check whether they are not only weakly sustainable. In order to check whether the sufficient condition is satisfied for strong sustainability of the current account deficits, that is, whether in equation (10), we estimated equation (10) by OLS and used the Wald restriction coefficient test to check if is statistically equal to 1. The results are reported in Table 5 and they indicate that the estimated coefficient is statistically significant at 1% level of significance for the four countries. This suggests that although imports and exports were found to be cointegrated for Central Africa Republic, Chad, Congo and Gabon; it seems that the current account deficits are only strong sustainable for Chad and weakly sustainable for Central Africa Republic, Congo and Gabon.

Following the test for sustainability of the current account deficits in CEMAC countries done on a country by country basis using threshold cointegration test of Enders and Siklos ⁷, we further complement the analysis by using Pedroni ^{18, 19} panel cointegration test to examine the sustainability of current account deficits in CEMAC area as a panel. Prior to that, panel unit root tests (IPS and Pesaran CADF) are conducted to detect the order of integration of the variables. Results in Table 6 suggest that, at 5% level of significance, exports and imports are non-stationary processes, integrated of order one, I (1).

The results in Table 7 show that both panel statistics and group mean panel statistics suggest that the null hypothesis of no cointegration can be rejected regardless of the deterministic components included (intercept or trend). There exists therefore a cointegration relationship between exports and imports in CEMAC countries, which would suggest that current account deficits are sustainable in CEMAC area. Since exports and imports are found to be cointegrated, panel dynamic OLS (DOLS) is used to estimate the long-run equation. The estimation results are presented in Table 7 and the restriction test rejects the null hypothesis that the cointegrating coefficient is equal to 1. This suggests that current account deficits are weakly sustainable in CEMAC countries as a panel.

However, although the use of panel data has a number of advantages over pure time series data, caution is needed when interpreting panel data results, especially when heterogeneity dimension among the cross-sections is not taken into account ⁵.

5. Concluding Remarks

This paper was intended to examine whether exports and imports are cointegrated for the Economic and Monetary Community of Central African countries, namely, Cameroon, Chad, Central African Republic, Congo, Gabon and Equatorial Guinea, in order to test for the sustainability of their current account deficits. Using TAR and momentum TAR models, the study adopted threshold cointegration test advanced by Enders and Siklos ⁷, allowing for asymmetric adjustment towards the long-run equilibrium.

The findings indicate that exports and imports are threshold cointegrated at 5% level for Central Africa Republic, Chad, Congo and Gabon. The current account deficits are only strong sustainable for Chad and weakly sustainable for Central Africa Republic, Congo and Gabon. Since the use of panel data has a number of merits over time-series data, panel cointegration test proposed by Pedroni ^{18, 19} was also applied to test for sustainability of current account deficits. The results from panel cointegration test showed that exports and imports are cointegrated and that the cointegrating coefficient is statistically equal less than 1, which would imply that current account deficits are weakly sustainable in CEMAC area. However, as Dumitrescu and Hurlin ⁵ point out, caution is needed when interpreting panel data results, especially when heterogeneity dimension among the cross-sections is not taken into account.

In summary, the findings suggest that Cameroon and Equatorial Guinea are in violation of their international budget constraints and should therefore put in place policies to reduce their current account deficits in order to regain their external stability. For the other country members of CEMAC, Central Africa Republic, Congo and Gabon, where sustainability was found to be weak, they should also implement policies to reinforce the sustainability of the current account deficits.

References