Fiscal Policy and Economic Cycles in Congo

The purpose of this paper is to respond on discretionary fiscal policy on economic cycles in Congo. A fiscal reaction function developed by Huart [1] was thus estimated over the 1989-2015 period. It comes out from the results that the fiscal policies carried out during this period were both counter-cyclical and pro-cyclical expansionist. This resulted in instability of the public debt and an accumulation of payment arrears.


Introduction
The world economies are subject to fluctuations that require the reaction of public authorities based ondiscretionary fiscal policies for economic stabilization purposes. In fact, following a period of unrest, instability and crises from the First World War to the aftermath of the Second World War, Keynesian discretionary fiscal policies proved successful until the late 1960s. Yet, the 1970s were marked by the international monetary system crisis 1 and the two oil shocks 2 challenged the Keynesian approach about public finances due to the unexpected consequences of the juxtaposition of inflation and structural unemployment 3 .
Thus, faced with the failures of Keynesian fiscal stimulus policies, supply-side theorists argued in favor of supply-side policies that lead to the reduction of public spending 4 and/or taxes.
Based on rational expectations, the authors of the new classical economy [2] 1 It was this crisis that had led to the end of the dollar-gold parity. 2 Respectively in 1973 and 1979. 3 In other words, Keynesian policies have been called into question because of the stagflation experienced by the advanced countries. [3] highlighted the inefficiency and ineffectiveness of discretionary policies and argued for regulation policies. The effectiveness of the rules is assessed in relation to their effects on the cost of borrowing, on the one hand, and budgetary discipline on the other. The question of the effectiveness of the rules in terms of the effects on the cost of borrowing refers to the sustainability of public finances.
On the other hand, the one concerning the effects of the rules on budgetary discipline consists of the compatibility of the budgetary rules aiming to the budgetary policy. Based on the objectives of fiscal policy according to the three functions theorized by Musgrave [4], this second concern arises in terms of a priori opposition between the rules whose objective is the sustainability of public finances and the function which assumes that in case of an economic shock, the deficitlevel as well as the debt level may increase. In these circumstances, the position about budgetary rules depends on the role assigned to fiscal policy in the area of cyclical stabilization. There are two types of theoretical orientation for this purpose.
Firstly, the theories lay emphasis on budgetary policy in stabilizing cyclical conditions and advocating flexible rules. Generally speaking, the proponents of these theories considerthe budgetary rules as an obstacle. Nevertheless, they argue for the compatibility of cyclical stabilization and fiscal discipline in such a way that the direction of discretionary fiscal policies is countercyclical regardless of cyclical conditions. Secondly, the theories consider fiscal policy inefficient regarding economic stabilization. Indeed, fiscal policy must be allocated to the sustainability of public finances and be limited to the automatic stabilizers, with monetary policy stabilizing the economy. However, these theoriesstipulate that cyclical fiscal policy may only be necessary in three cases [5] [6]. The first case is about a situation of deflation risk 5 , i.e., when the economy is in a liquidity trap situation.
The second case concerns a situation wherein monetary policy undermines the credibility of the central bank. The third case deals with a fixed parity situation, with capital mobility where monetary policy is mobilized for the objective of the exchange ratestability.
Notwithstanding these theoretical orientations, the financial and economic crisis that occurred at the end of 2007, coupled with the effects of fluctuations in oil prices, has revived the debate on the effectiveness of fiscal stabilization policy.
Thus, during the last decade, several studies carried out within the fiscal policy framework have focused on the reaction of discretionary fiscal policy on economic cycles.
Congo, a member of the CEMAC 6 , is subject to the budgetary rules in force in the countries of this zone, undergoing economic cycles like any other countries, hence it is pertinent to analyze the fiscal policies implemented by its different governments in connection to economic cycles. In other words, it is worth 5 The situation where nominal interest rates and inflation are close to null. 6 Economic and Monetary Community of Central African States. In addition to counter cyclical and acyclic fiscal policies, the theoretical literature also identifies procyclical fiscal policy. Indeed, a fiscal policy is procyclical, if the public authorities increase (decrease) public expenditure and decrease (increase) the taxes in theexpansion phase (recession). Such a fiscal policy strengthens the amplitude of the economic cycle which is detrimental in terms of inflation and employment.
Finally, proponents of the threshold effects of fiscal policy suggest the coexistence of different fiscal regimes contingent on public debt. In this respect, three schemes can be distinguished. Firstly, when economic slowdown with debt is below the sustainability threshold (or the level of public spending is below the optimal size of the state), expansionary fiscal counter-cyclical policies must be applied in order to reinforce the effects of automatic stabilizers. Secondly, beyond the optimal size of the State, in the event of economic recovery, countercyclical restrictive fiscal policies must be pursued to reduce the level of public expenditure to the optimal level [7] to which acyclic budgetary policies will apply. It is in this context that unconventional fiscal policy tenants suggest that during a crisis, discretionary stimulus measures should be targeted, timely and temporary [8] to the extent that when the crisis is over, the opposite steps must be taken. Thirdly, with a debt ratio above the threshold of sustainability followed by an economic slowdown, a budgetary contraction would have non-Keynesian or anti-Keynesian effects. In other words, a fiscal contraction (i.e. 9 For Barro, the acyclical fiscal policy, which is an optimal fiscal policy. nature of fiscal policy differs from one country to another in OECD countries.
Indeed, it is countercyclical in advanced countries, whereas it is likely to be procyclical in countries whose production is more volatile and where political power is more dispersed. However, at the disaggregated level of public spending, the behavior of transfers and interest on debt is countercyclical, while that of current expenditure and public investment is procyclical.

The Budgetary Policy Rules in Force in the Congo
The governments to practice counter-cyclical fiscal policies [32] but rather leads to a procyclical bias in public spending [19]. In general, Congo met this criterion indicator that assesses the financing difficulties that a country may face in its economic development. Indeed, internal arrears are likely to increase inflationary pressures due to the anticipatory effects of economic agents [33]. The infla- 13 These are the first criteria to which must be added the second-tier criteria: Wage/tax revenue < 35%; Investments financed from own resources > 20%; Current account balance/GDP > 5%; Tax rate > 17%; Foreign exchange rate of foreign exchange assets > 20%. 14 This balance, again called the basic primary balance, excludes interest on debt and investment expenditure financed by own resources 15 According to the IMF's June 2017 mission, Congo's public debt would be just over 90% of GDP, while the HIPC debt relief had reduced the debt ratio to 15% of GDP. Since more than 2013, the aggregate of arrears in domestic wage payments amounts to more than 15 billion FCFA. 16 Such a mechanism has been put in place in Poland. The debt ceiling is set at 60% of GDP, with two alert thresholds set at 50% to 55% of GDP, respectively. Overall, these rules have the disadvantage of not fixing a structural deficit rule as a percentage of GDP in order to restore the medium-term budget balance which should be able to face unexpected fluctuations or economic shocks. But since Congo is a country whose public finances solely depend on oil exports and its price evolution on the international market, the conventional methods used 18 to anchor fiscal policy seem inadequate to its economy. Taking this particular situation into consideration, the anchoring of fiscal policy in Congo must combine the objective of fiscal sustainability with greater flexibility in public spending to respond to shocks and to maintain the cyclical role of public finance of the State. By so doing, the rules currently in force should at least be complemented by three rules: a fiscal rule based on the oil price, i.e., a structural primary balance rule with smoothing of oil prices; a structural income rule for non-oil revenues, i.e. a primary balance rule excluding natural resources, and a rule for a public spending growth 19 .
First of all, the rule based on the oil price 20 can be used to smooth the oil revenues and the planned public spending. When real prices are above (lower) the budgeted, revenues are higher (lower) than revenue forecasts, resulting in a surplus (a deficit) that gives rise to accumulation of the stabilization fund. It then 17 Since more than 2013, the aggregate of arrears in domestic wage payments amounts to more than 15 billion FCFA. 18 This is a combination of flow-related rules (in the form of a budgetary target, often adjusted to the cycle) and stocks (the ratio of public debt to GDP). 19 These are the IMF's recommendations to countries rich in natural resources. 20 This is a rule of the structural primary balance with price smoothing. With respect to the non-oil structural revenue rule, excluding the cyclical component of output 21 , it aims to reduce revenue volatility and the pro-cyclical nature of the budget. Indeed, non-oil structural revenues can be calculated by adjusting non-oil revenues to the ratio of potential output to real output.
Finally, the rule of public expenditure growth permits to limit the procyclicality of fiscal policy. It can also be used when there is insufficient absorption capacity 22 . This last rule is generally combined with a price-based rule. All these two rules can be elaborated only in countries rich in natural resources whose exhaus- are not yet applied in Congo, our assessment of fiscal policy will focus on the primary structural balance which includes both primary non-oil and non-oil production, as well as total production (oil and non-oil).

Model for Determining the Cyclicality of Fiscal Policy
To deal with the problem raised, we are going to test the relationship between the change in the primary structural budget balance and the change in the output gap.

Variables of the Model
Three variables are used to determine the cyclicality of fiscal policy in Congo: the change in the primary structural budget balance (SBSP), the change in the output gap (PF) and the public debt-to-GDP ratio due to its effects on the primary structural budget balance.
. Y: produced production; 21 This is the rule for the structural primary balance excluding natural resources. This is a rule that applies to countries where the depletion of natural resources takes place within a relatively short time horizon. 22 It can be used for overheating or high current deficits. 23 The structural budget balance is deduced from Taylor's [5] fiscal policy rule that There is underemployment and therefore unused production capacity.  y y − < , the cyclical balance is neutral when y y = . It follows, finally, that a discretionary easing in a cyclical downturnperiod is a countercyclical policy, and in a cyclical expansionperiod is a procyclical policy. On the other hand, a discretionary tightening in a period of cyclical downturn is a pro-cyclical policy, and in a period of cyclical expansion, it is a countercyclical policy. (SBSP t-1 ) among the explanatory variables of the current structural primary budget balance (SBSP t ) reflects the influence of the initial budgetary conditions on the budgetary decisions of a given time in particular the inertia observed in the evolution of fiscal policy variables due to implementation delays or measures that are difficult to reverse. The sign of the coefficient β 1 is expected to be positive.For the parameter β 2 attached to the output variance (EP) variable, it denotes a contra-cyclical (pro-cyclical) discretionary fiscal policy if it is positive (negative). As for the public debt-to-GDP ratio, it represents a "discipline" effect: a positive β 3 parameter indicates a motive for stabilizing the debt under fiscal policy (a higher debt-to-GDP ratio entails budgetary efforts to improve the primary structural budget balance).

Results of the Estimate and Their Implications
Thefirst step is to present the results of the estimation, and then to analyze their implications (Table 1).

The Results of the Estimation
The data cover the period from 1989 to 2015 and their sources are presented in Table A1 in the Appendix. Since the period is short (less than 30 observations), the technique used is quarterly to increase the sample size and have more consistent estimators. This technique is recommended by some international institutions like the IMF 25 .
We first evaluated the value of the coefficient ε (see Table A2 in the Appendix) to calculate the structural budget balance. The value of ε is 0.098. A decrease of one point in the output gap leads to a decrease in the cyclical surplus (in-

Interpretation of Results
The  Thus, the Congolese State would benefit from a counter-cyclical expansionary fiscal policy (increase in public spending and fall in public revenues during a downturn) and a counter-cyclical restrictive fiscal policy (falling public spending and rising government revenues during a recovery period) in order to avoid the risk of an increase in public debt that could lead to the unsustainability of public finances. Indeed, a countercyclical restrictive fiscal policy would enable to identify the budgetary surpluses needed to reduce public debt. The fiscal ad-justment in Congo during the recovery period should be based mainly on the public expenditure component as compared to revenue.
Moreover, since the effectiveness of automatic stabilizers is low, it is recommended to implement active public policies. In other words, the Congolese government should make greater use of the structural component of fiscal policy as an instrument of economic policy.
Finally, Congo's economy depends heavily on a natural resource, oil, the countercyclical nature of fiscal policy should be strengthened by the implementation of three budgetary rules: a structural primary balance rule with smoothing of oil prices; a structural income rule for non-oil revenues, anda public expenditure growth rule. However, the credibility and effectiveness of these rules should require the combination of the establishment of a sovereign stabilization fund to limit spending and smooth public revenues which should allow an active counter-cyclical policy. Moreover, since the output gap is negative, Congo is experiencing involuntary unemployment in the sense of Keynes. As a matter of fact, the State should intervene to improve the functioning of markets and the efficiency of public investment so that private production is more important and national production reaches its potential level. In other words, the State should also consider the influence of structural policies that can play an important role in the evolution and capacity of the economy to withstand shocks. The Congolese State should intervene within the framework of the functions of resource allocation and income redistribution in order to promote the development of the additional production capacity needed to diversify the economy. Indeed, a diversified economy is more resilient to shocks.

Conclusion
The objective of this research was to determine the cyclicality of Congo's fiscal policy from 1989 to 2015. The results of the estimation of a fiscal reaction function showed that the fiscal policy implemented by different successive governments in Congo have been both counter-cyclical expansionist and procyclical expansionist leading to the instability of the public debt that could lead to the unsustainability of public finances. The hypothesis of the pro-cyclical restrictive fiscal policy was rejected in favor of the counter-cyclical restrictive fiscal policy. On the other hand, the countercyclical expansionist fiscal policy has been confirmed. The non-cumulative arrears and debt thresholds were not met due to pro-cyclical expansionary fiscal policies. The Congolese authorities would therefore benefit from counter-cyclical fiscal policies both during periods of economic downturn and resumption so that the current budgetary rules can be respected.