ABSTRACT IN ENGLISH:
This paper analyzes theoretically and empirically the interactions between fiscal and external sector variables with a reference to the Moroccan case. The paper seems to be innovative on theoretical and empirical fronts in the extent that it deeply examines such interactions on the theoretical level and attempts to empirically validate them through multivariate reduced-form as well as structural multivariate models. The estimated structural multivariate models aim at analyzing effects of structural shocks as generated by fiscal and external sector variables.
First, the paper examines the theoretical foundations and previous studies regarding effects of fiscal variables on external sector indicators. It focuses then on the formulation and empirical validation of additional theoretical hypotheses on the opposite links between the two sets of variables keeping in mind that such opposite linkages are largely under-explored in the existing theoretical literature.
To examine how the various structural shocks impact the variables under study in an interactive way, we have estimated a structural -var-(SVAR) model. Within this framework, we have used two complementary analytical tools: the response --function--s to shocks and the variance decomposition of forecast errors. While the response --function--s reveal that fiscal surpluses in Morocco are sensitive to fiscal as well external shocks (especially terms of trade and foreign demand-induced shocks), the variance decomposition of forecast errors shows a relatively pronounced predominance of fiscal shocks in explaining fluctuations in fiscal balances as well as external current account surpluses.
Key Words: Fiscal shocks, external shocks, interactions, SVAR.
ABSTRACT IN FRENCH:
Ce papier se propose de mener une analyse théorique et empirique sur les interactions entre les variables du secteur externe et celles des finances publiques en se référant au cas du Maroc. Le papier se veut novateur au double niveau théorique et empirique en ce sens qu’il examine théoriquement les interrelations en profondeur qu’entretiennent les variables budgétaires et celles du secteur externe tout en essayant de les valider empiriquement à travers des modèles multivariés aussi bien à forme réduite que structurels. L’estimation de modèles multivariés structurels vise essentiellement à analyser les effets des chocs structurels générés par les variables budgétaires et celles du secteur externe.
De prime abord, le papier examine les fondements théoriques et les études empiriques concernant les effets des variables budgétaires sur les indicateurs du secteur externe. Il se penche ensuite sur la formulation et la validation empirique d’hypothèses théoriques additionnelles sur les liens entre ce groupe de variables dans le sens opposé en gardant à l’esprit que les relations dans cette --dir--ection sont largement sous-explorées dans la littérature théorique existante.
En vue d’examiner les effets des différents chocs structurels sur les variables étudiées sous une optique interactive, nous avons estimé un modèle multivarié structurel de type vectoriel autorégressif structurel (SVAR). Dans ce cadre, nous avons utilisé deux outils d’analyse complémentaires : les fonctions de réponse aux chocs structurels et la décomposition de la variance des erreurs de prévision. La première analyse révèle que les soldes budgétaires au Maroc sont sensibles aux chocs aussi bien budgétaires qu’externes (notamment les chocs induits par les termes de l’échange et la demande étrangère). Quant aux résultats de la décomposition de la variance des erreurs de prévision, ils montrent une prédominance relativement prononcée des chocs budgétaires dans l’explication des fluctuations des soldes budgétaires et des soldes courants de la balance des paiements.
Mots clés: choc budgétaire, choc externe, interactions, SVAR
It is commonly admitted at the present that fiscal policy crucially matters, especially in the developing World. Even though they don’t accept a strict economic intervention of the State, classical economists tolerate the maintaining of a minimal State in the sense that governments may intervene only to produce and allocate certain goods which cannot be provided by the private sector. By contrast, Keynesian economists consider fiscal policy as a powerful tool to boost economic recovery. Fiscal policy has to play the important roles of economic stabilization and income redistribution, in addition to efficient public service delivery and allocation of available resources. As Tinbergen argues, fiscal and tax policies are seen to be efficient in driving economic activity in the extent that they help to boost economic growth´-or-to slowdown the rhythm of activity in order to avoid overheating With the accelerating openness of domestic economies, economic policy has to play a new role that is the equilibrium of external accounts, especially the balance of payments’ equilibrium. The openness of domestic economies enlarges interactions with partner countries, then endangering exogenous shocks gradually exacerbated through the interdependence of economies and economic policies worldwide.
Since early 1980s, Morocco is engaged in a process of economic liberalization at a time when the domestic economy was considered as liberally unachieved. Following devoted efforts in macroeconomic stabilization and structural reforms since 1983, Morocco started a new process of free trade agreements with a variety of countries such as the European --union--, the Aga--dir-- Agreement Countries (Tunisia, Egypt and Jordan), the USA and Turkey. As most of Morocco’s partners are more competitive, such openness is viewed as a challenge for the domestic economy. In addition to the fact that the latter is a “small open economy” which is very sensitive to fluctuations of external sector variables, it is also subjected to structural shortcomings with respect to the weak competitiveness of the productive system with its implications in terms of external accounts’ equilibria.
In addition, during recent years, following the economic and financial crisis in 2008 , and political tensions in the MENA Region, several indicators of the domestic economy, notably those tied to the external sector and public finance, have experienced sharp regressions. Indeed, World real demand for Moroccan merchandises experienced a fall of about 3.56 percent in 2009 and 3.7 percent in 2012 (Ministry of Finance, 2013). Terms of trade also experienced a fall of about 1.64 percent in 2011 and 5.4 percent in 2012. Moreover, during recent years, fiscal and external imbalances have become more pronounced, even surpassing the fixed IMF and World Bank norms. Fiscal and external deficits have almost followed the same trend with a mutual evolution. For instance, in 2012, fiscal and current account deficits amounted to 7.3 and 10 percentage points of GDP respectively. These statistical figures motivate a plausible questioning about possible interactions between fiscal and external sector variables. In what follows, we present our major research problems before outlining our general and specific objectives as well as our overall and specific research hypotheses.
Our research problems are generally tied to the interaction between fiscal and external sector variables. Fiscal variables are defined here as fiscal surpluses and public revenue and expenditures. As far as external sector variables are concerned, they are selected as foreign demand, terms of trade, the balance of payments’ current surplus and the real effective exchange rate.
In general, our main research problems may be outlined as follows:
- Primo: If the existing economic theory and certain empirical works focus on the relationship going from fiscal to external sector variables, especially those which are relatively controllable , what is about the opposite linkages? Within this framework, the causality going from external sector variables to those of public finance has received less attention on theoretical as well as empirical fronts. Moreover, most of the existing empirical studies about the impact of fiscal policy on external balances, particularly in the Moroccan case, have used only reduced-form models which are seen to be unable to better inform about the complexity of interrelations between the two groups of variables.
- Segundo: if external sector variables affect public finance indicators in Morocco, how can we measure the possible contributions of external shocks to budget fluctuations? What would be the respective effects of fiscal and external sector shocks in an interactive way, using a circular reasoning methodology?
- Trio: what may be the suitable methodological tools to adopt in order to confirm´-or-infirm the above-mentioned interactions and analyze the transmission of shocks from external sector variables to those of public finance and vice-versa?
These are important questions engaging researchers and policymakers as in a “bridging research and policy” perspective. Our attempts to answer these various research questions rely on a variety of theoretical paradigms and methodological tools devoted to respect the triangulation principle as known in social sciences.
Our research hypotheses may be formulated as follows
• In line with the old economic theory debate, external sector variables would affect fiscal indicators--;--
• Additional theoretical hypotheses may predict that external sector variables would influence public finance variables--;--
• The best methodological tool to test such interactions and analyze the transmission of external sector shocks toward public finance variables and vice-versa is the structural multivariate modeling: Structural Vector Auto-Regressive Model (SVAR), Structural Vector Error Correction Model (SVECM), etc.
• In spite of the robustness of structural multivariate modeling, reduced-form multivariate models (error-correction models, etc.) with short and long-run causality tests would be useful to test our additional theoretical hypotheses about the causality going from external sector to fiscal variables as well as to better impose necessary restrictions for identification and estimation of our structural multivariate models.
In line with our research problems and hypotheses as above outlined, this paper aspires to conducting a theoretical and empirical analysis of interrelations between external sector and fiscal variables in an interactive way. More specifically, our objective is to analyze and measure the effects of fiscal shocks on external sector variables and vice-versa. We aim in definitive at conducting relaible empirical analysis for decisionmaking purposes.
To test our research hypotheses, to bring answers to our research problems and therefore to achieve the expected objectives, the whole paper to be published in an international indexed journal is organized as follows:
- Section 1 presents a critical review of the theoretical and empirical literature about linkages between fiscal and external sector variables along with the formulation of testable additional theoretical hypotheses regarding the causality going from external sector variables to those of public finance--;--
- Section 2 examines the methodological framework to adopt in order to conduct an empirical investigation of the interactions between the two groups of variables, with a focus on the usefulness of structural multivariate modeling--;--
- Section 3 applies the examined methodological framework in order to analyze the interactions which may exist between external sector and fiscal variables in the particular case of Morocco--;--
- Finally, section 4 formulates some policy implications and concludes.
Remember that the studied external variables are real foreign demand, the real effective exchange rate, terms of trade and the current account surplus while domestic variables are defined as real GDP and fiscal surplus as proportions to GDP.
To better benefit from major information channeled through two variants of our S-var-model with relative prices alternatively measured through the real effective exchange rate and terms of trade, it is essential to compute the response --function--s of the studied variables to the various structural shocks as identified, and to determine the decomposition of the variance of forecast errors. The analysis based on the response --function--s allows us to know the sign and the nature of impact (persistent´-or-transitory) of every structural shock on the various components of the -var-system at a chosen time horizon. Regarding the analysis of the forecast error decomposition, it permits to determine the contribution of each structural shock to fluctuations of the studied variables at a selected time horizon.
First, we present and interpret the empirical results related to the effects of fiscal shocks on the ratio to GDP of the balance of payments current surplus and the real effective exchange rate. Second, we present and interpret empirical results regarding the effects of shocks due to external sector variables (foreign demand and terms of trade) on fiscal surpluses as a key public policy variable.
IMPACT OF FISCAL SHOCKS ON EXTERNAL SECTOR VARIABLES:
In what follows, we study the impact of fiscal shocks on current surpluses before analyzing the effect of these shocks on real effective exchange rates.
IMPACT OF FISCAL SHOCKS ON CURRENT SURPLUSES:
Regarding the sign and nature of the impact of fiscal shocks on current surpluses, empirical results based on impulse response --function--s show that a positive fiscal policy shock leads immediately to improving ratios to GDP of current surpluses (in first difference). This impact is seen to be relatively persistent since it tends to meet its long term level only at a time horizon of two years (eight quarters). This empirical result is consistent with economic theory predictions, especially the well-known twin deficits hypothesis.
Concerning the contribution of fiscal shocks to current account fluctuations, empirical results indicate that the variability of the ratio to GDP of the current surplus is due for nearly 49 percent to fiscal shocks, 37 percent to its own shocks and 14 percent to other structural shocks, for a time horizon varying from the first to the twentieth quarter. The contribution of fiscal policy shocks to current account fluctuations is therefore seen to be relatively strong. These empirical results explain that external imbalances are strongly sensitive to fluctuations of public finance fluctuations, coinciding thereby with the above-interpreted results from impulse analysis.
IMPACT OF FISCAL SHOCKS ON THE REAL EFFECTIVE EXCHANGE RATE:
It is important to outline the reaction of real effective exchange rate (TCER) to a fiscal policy shock. Empirical results reveal that a positive fiscal shock (improving fiscal surplus as proportion to GDP) induces immediately depreciating real effective exchange rates (TCER). Such impact reaches its optimum at the second and sixth quarters.
Starting from the 10th quarter, such effect converges slowly towards its long term level, which is reached at a time horizon of five years (twenty quarters). These empirical results are consistent with economic theory predictions and meet results found in some previous empirical works such those done by Brahim Mansouri in 2003 and 2013--;-- and Blot in 2005.
Analysis of the variance of forecast error decomposition yields empirical results revealing that fluctuations of the real effective exchange rate are almost due to their own shocks, in the short as well as long terms. Over a time horizon varying from the first to the twentieth quarters, the variability of the real effective exchange rate is due for about 75 percent to autonomous exchange rate variations, 14 percent to fiscal shocks and 11 percent for other identified structural shocks.
The relative weakness of the fiscal shock contribution to the real effective exchange rate fluctuations would likely be due to the fact that decreasing domestic prices, initially due to improving fiscal surpluses and then improving current account surpluses, would be significantly offset by a similar decrease of prices in the main trading partners, especially the European --union-- countries. Indeed, economists and decisionmakers agree that inflation in Morocco is seen as comparable to inflation levels prevailing in the Euro-land.
EFFECTS OF EXTERNAL SECTOR SHOCKS ON FISCAL SURPLUSES:
In the following developments, we analyze the reactions of the ratio to GDP of fiscal surpluses (in first difference) to foreign demand shocks and determine the contribution of such shocks to fluctuations of this budget variable. We then analyze the reactions of fiscal surpluses to shocks tied to terms of trade and estimate the contribution of these shocks to the same variable.
IMPACT OF FOREIGN DEMAND SHOCKS ON FISCAL DEFICITS:
It is interesting to outline the reaction of fiscal surpluses as proportions to GDP to a foreign demand shock. Results for impulse response --function--s reveal that a positive foreign demand shock exerts an immediate impact on the ratio to GDP of the budget surplus (in first difference).
During the first quarter, the shock impact on fiscal deficits permeated by foreign demand is negative--;-- but, in terms of magnitude, such impact turns to be almost zero. Starting from the second quarter, the impact of such shock becomes positive and grows gradually to reach its maximum at the end of the tenth quarter. The impact is relatively small but persistent since it regains its long term level only starting from a time horizon of four years and a half (eighteen quarters).
To deeply analyze the effect of external shocks on public finance, we examine in what follows, the sources of public deficit fluctuations through a decomposition of the forecast error variance for the ratio to GDP of fiscal surplus (in first difference). Empirical results reveal for a time horizon varying from the first to the twentieth quarters that the variability of the fiscal surplus ratio to GDP (in first difference) is due for only about 2 percent to foreign demand shocks against 69 percent for own fiscal shocks and 29 percent for other structural shocks. This suggests that the endangering fiscal deficits experienced in Morocco during the studied period are largely due to a domestic structural component tied to decisionmakers’ discretionary measures in fiscal policy rather than to induced effects from the economic environment (for analytical and empirical details on this issue, see Mansouri’s works).
The weak impact of foreign demand shocks on fiscal surpluses as proportions to GDP would be associated with the fact that the positive impact of foreign demand on the ratio to GDP of the balance of payments current surplus is not relatively strong. Well, since the positive impact of fiscal surpluses on fiscal balance is not relatively robust, a positive foreign demand shock would likely improve fiscal surpluses only weakly. More accurately, the weakness of such impact would be probably associated with the fact that improving current surpluses, due to increasing foreign demand, raise public revenue but also dope public spending so as, in definitive, the net effect would be only a small improvement in the fiscal stance.
IMPACT OF TERMS OF TRADE SHOCKS ON THE FISCAL STANCE:
Results from impulse response --function--s show that a shock associated with terms of trade exerts an immediate and positive impact on the ratio to GDP of fiscal surplus. This impact is maximum at the first quarter and then decreases to reach its minimum at the end of the eighth quarter. Then, the impact grows slowly to become positive at the thirteenth quarter and regain its equilibrium level only at a time horizon of five years (twenty quarters). Such long-run positive impact would be likely due to the dynamics of comparative advantages that Morocco has experienced during the recent years, especially with the reinforcing production and export of manufactured goods in the framework of the so-called “New World Professions of Morocco” (Nouveaux Métiers Mondiaux du Maroc): automotive industry, aeronautics, electronics, etc.
Analyzing the decomposition of the forecast error variance for the ratio to GDP of fiscal surplus, empirical results reveal that terms of trade shocks remain the major external cause of fiscal surplus fluctuations. Indeed, the contribution of such external shocks to fiscal surplus fluctuations (in percentage of GDP) is seen to be around 15 percent of the total contribution of the aggregate structural shocks. Nevertheless, such contribution to variations of fiscal surpluses as proportions to GDP turns to be weaker in comparison with the contribution of autonomous fiscal shocks. This would be likely due to the fact that terms of trade, while improving public revenue, also boost public expenditures in slightly similar proportions.