Cemac: Application of a Debt-Investment-Growth Model1
The fall in the price of crude oil since July of 2014 has an important impact on CEMAC, since five of the six member countries are net oil exporters. An application of a debt-investment-growth (DIG) model confirms the importance of the shock, and shows that robust growth and sustainable macroeconomic balances require policy adjustments. These include measures to enhance the impact of public expenditure, to scale back said expenditure, and to generate additional revenue. The application of the DIG model also points to the difficulty in maintaining an effective stock of public capital when public expenditure must be reined in.
1. This paper assesses the sustainability of public investment, growth, and public debt in the context of the precipitous fall in oil prices since mid-2014. It builds on the chapter entitled “Public investment scaling-up, growth, and debt dynamics” (IMF 2014), itself based on the DIG model developed by Buffie, Berg, and Zana (2012; Box 1), with an update of key exogenous variables (Box 2), a new analysis of the impact of policy adjustments on key economic variables (Box 3), and a discussion of policy options.
2. The oil price slump has a dramatic effect on economic and fiscal prospects for CEMAC, with looming challenges in the medium term. Growth is now slated to fall by up to 2 percentage points in 2015 relative to medium-term projections from a year ago (Figure 1); and oil revenue is projected to fall by more than 10 percentage points of the new, more modest GDP projections. This change in circumstances raises issues of fiscal and external sustainability that may be addressed, in part, by policy adjustments.
Figure 1.CEMAC: Selected Economic Indicators, 2014–15
Sources: authorities’ data; and IMF staff estimates and projections.
Box 1.CEMAC: Key Features of the Model
The debt-investment-growth (DIG) model is a computable general equilibrium model that offers an alternative to the classic “financial programming” model. It is inherently different in that growth is an endogenous variable, and thus offers interesting insights on the dynamics between investment, growth, and debt after initial conditions are specified.
The model features three production sectors: (i) the oil sector; (ii) the sector producing non-oil internationally traded goods; and (iii) the non-traded goods sector. The oil sector is modeled as a yearly endowment that is exported, and the proceeds of which are shared between the government, households, and extracting firms. The other two sectors are modeled as a Cobb-Douglas production function involving public capital, private capital, and labor.
Households comprise savers and non-savers. Savers can invest in government bonds, and may borrow at significant rates in an international financial market.
Governments can decide to spend in the form of transfers to households or public investment, and must pay interest on borrowings. Their resources comprise external concessional borrowing, external non-concessional borrowing, domestic borrowing, grants, and a consumption tax.
Public capital is expanded by yearly capital expenditure, but is discounted by an efficiency factor that accounts for leakages of public resources.
Box 2.CEMAC: Initial Conditions
Initial conditions for the model in “year zero” mirror closely CEMAC’s circumstances in 2013. The initial levels of some of the important variables that will determine sustainability are either CEMAC-specific, or calibrated to low-income countries. They are as follows:
an initial public debt level of 30 percent of GDP, split equally between domestic debt, external concessional debt, and external commercial debt;
interest rates on domestic debt and external non-concessional debt are 6 percent and 8 percent, respectively;
foreign grants are limited to 1 percent of GDP;
the consumption tax rate is 15 percent;
public investment spending is 12 percent of GDP;
the public investment efficiency level is 40 percent;
the financial return on public capital is 20 percent;
the ratio of non-savers to savers is 1.5;
capital’s share in the value added of the tradable sector is 40 percent;
capital’s share in the value added of the non-tradable sector is 55 percent;
value added in the non-tradable sector is 49.4 percent; and
the ratio of imports-to-GDP is 30.5 percent.
Box 3.CEMAC: The Model’s Output Variables
The various scenarios that follow are each accompanied by eight graphs that illustrate the profile of relevant variables. These variables have been selected to highlight the important dimensions of sustainability (fiscal and external), the trade-offs between sustainability and poverty alleviation, and meaningful transmission mechanisms by which changes in assumptions affect policy objectives. The variables are summarily discussed below.
|“Scaling Up” - displays the values of investment spending for all CEMAC governments, as a percentage of CEMAC GDP.||“Tax Rate” – shows the actual percentage of the value of consumption that is captured by governments in the form of a tax.|
|“Oil Revenues” – represents the value of revenues from oil for all CEMAC governments, as a percentage of CEMAC GDP.||“Public Effective Capital” - shows the increase over year zero of the actual public capital formed from public investment spending.|
|“Private Capital” – shows the increase over year zero of the actual private capital formed.||“Real per Capita GDP Growth” - illustrates year-on-year increases of GDP per capita (i.e., real GDP growth net of population growth).|
|“Current Account Deficit” – refers to the external current account deficit in percent of GDP, an important indicator of external sustainability.||“Total Public Debt” – is the main indicator of fiscal sustainability; it is preferred to the fiscal deficit over longer projection periods because it factors in amortization patterns.|
B. Baseline Scenarios—A Shift in the Outlook
The Original Baseline Scenario: “What the future looked like a year ago”
3. The original baseline scenario is a close approximation of the scenario used in the appendix to the 2014 CEMAC Country Report (IMF 2014). It is predicated on five important trends: (i) a profile for public investment spending that spikes at 16 percent of GDP in 2014 and gradually declines to 12 percent by 2020; (ii) debt financing of all public investment spending, less 4 percentage points of GDP; (iii) a flat consumption tax rate of 15 percent; (iv) stable current spending relative to GDP; and (v) crude oil revenue reflecting a gradual volume decrease, and the June 2014 IMF World Economic Outlook (WEO) price projections, when the price of oil was expected to ebb only slightly from its high level of more than US$110 per barrel.
4. Until mid-2014, the macroeconomic outlook for the region was sustainable (Figure 2). Growth per capita was to pick up sharply with the scaling up of investment spending, and as private capital formation increased, growth per capita was to continue at a slightly lesser, but still commendable, rate in the long run. In the medium term (up to 2020; year 7 on the horizontal axis), overall public debt was to increase only gradually to under 50 percent of GDP, which is less than the CEMAC convergence ceiling of 70 percent, despite an assumption of consistently low consumption taxes. Almost as important, the effectiveness of new public capital was to increase sharply and remain strong over time, despite a small dip between years 10 and 14.
Figure 2.CEMAC: Original Baseline Scenario Projections
The Current Baseline Scenario: “The New Reality”
5. The current baseline scenario retains all the assumptions from the original baseline scenario, with the exception of oil revenue, which is now set to fall by 40 percent for any given year relative to the baseline scenario, on account of the international oil price shock (Figure 3). This adjustment is consistent with the oil price projections from the WEO of April 2015. The fall in oil revenue is completely offset by external non-concessional borrowing.
Figure 3.CEMAC: New Baseline Scenario
6. In the absence of any policy adjustment, the fall in oil revenue causes the macroeconomic outlook to become unsustainable. Total public debt increases faster, especially in the medium term, when it exceeds 50 percent of GDP within 5 years, and the external current account deficit evolves in the same manner. However, the shift of the budget financing burden from oil revenue to external debt releases domestic resources that allow for private capital to accumulate faster, and raises growth per capita marginally over time.
C. Reform Scenarios—Mapping a Path to Sustainable Development
Reform Scenario #1: “Improved Public Investment Efficiency”
7. The government has an array of policy tools to improve macroeconomic sustainability. One method is to improve the efficiency of public spending with the help of new assessment tools.2 Under this scenario, the efficiency of spending increases from 40 percent to 80 percent,3 a level that is common in faster developing economies (Figure 4).
Figure 4.CEMAC: Reform Scenario #1: “Improved Public Investment Efficiency”
8. This reform affects growth more than fiscal balances. Effective public capital accumulation almost doubles, which contributes to a marginal improvement in growth per capita; but it does not affect the profile of public debt meaningfully, especially in the medium term, since the tax rate is set at 15 percent by design; and the tax base (i.e., consumption) is undermined by a higher propensity to allocate resources to private capital, which is stimulated by higher levels of public capital.
Reform Scenario #2: “Improved Efficiency and Return on Public Investment”
9. Governments can choose to improve income from new public infrastructure, which is originally set at a low level of 20 percent in this model. A new higher return of 40 percent could come from selecting public investments that can generate a cash flow, and charging higher user fees for these investments, such as roads, bridges, dams, or port concessions (Figure 5).
Figure 5.CEMAC: Reform Scenario #2: “Improved Efficiency and Return on Public Investment”
10. The main effect of this policy adjustment is, as with the previous reform, to enhance growth more than fiscal balances. Public debt accumulation in this scenario is somewhat lower. A higher return on public investment gradually increases the effectiveness of public capital, thus helping to sustain the level of productive public capital. Debt sustainability remains doubtful for the same reason as in the previous scenario: the tax rate is broadly constant, and the tax base does not grow as fast as the overall economy, because faster private capital formation is crowding out consumption.
Reform Scenario #3: “Improved Efficiency and Return on Public Investment, with Lower Scaling Up of Investment”
11. This scenario is the first to introduce an adjustment to a variable that affects fiscal sustainability directly. The previous reform scenario is now complemented with the assumption of a limited scaling up in the near term. The increase in public investment is divided by two (e.g., for 2014, public investment is lowered from 16 percent of GDP to 14 percent of GDP starting from a level of 12 percent of GDP the year before). It is worth noting that this remains a “scaling up” scenario, with public investment at consistently high levels of more than 10 percent of GDP (Figure 6).
Figure 6.CEMAC: Reform Scenario #3: “Improved Efficiency and Return for Public Investment and Low Scaling Up”
12. This scenario has a strong positive effect on debt sustainability in the medium term. Total public debt reaches only 40 percent of GDP by 2019. In the long term, however, debt dynamics are again dominated by the severe shortfall in oil revenue. Real GDP growth per capita is lower during the scaling up period owing to lower public expenditure, but it maintains a favorable, consistently increasing profile beyond the years of the scaling up.
Reform Scenario #4: “Improved Efficiency and Return on Public Investment, with a Gradual Scaling Down of Investment and a Marginal Tax rate Increase”
13. This scenario includes two new adjustments: (i) a permanent dip in investment to 11 percent of GDP after a more limited scaling up; and (ii) a gradual increase in the rate of consumption taxes (from 15 percent to 16 percent; Figure 7).
Figure 7.CEMAC: Reform Scenario #4: “Improved Efficiency and Return for Public Investment with a Gradual Scaling Down and a Marginal Tax Rate Increase”
14. The additional adjustments have a dramatic and favorable effect on sustainability. The rise in the tax rate has a more significant effect over time, when the oil revenue shortfall becomes more pronounced, thus offsetting much of the latter’s effect on debt accumulation. In this scenario, however, the higher tax burden causes a noticeable crowding out of private investment; and lower public investment undermines the marginal production of private capital. Both of these developments dampen economic growth per capita over time. Importantly also, effective public capital dips slightly over time.
D. Conclusion and Policy Options
15. The various iterations of the DIG model for CEMAC show that certain policy tools have a significant direct effect on growth per capita, with indirect consequences on fiscal balances. The most important among these variables is the efficiency of public investment, which, when doubled, prompts a jump in the rate of growth per capita by around one percentage point within five years. Other variables have a strong direct impact on fiscal balances, with a knock-on effect on growth per capita. These variables are the magnitude of the scaling up of investment, and the consumption tax rate. It is worth flagging that a lower level of volatility in public investment (i.e., a lesser “hump” in the scaling up profile) also leads to a more stable stock of effective public capital.
16. CEMAC governments may exercise a degree of control over these policy variables. Improving the efficiency of public spending points to reform areas that include public procurement and financial management, which are currently weak in CEMAC. Moreover, efficiency improvements are as much a matter of project selection as process administration. Coherent project selection in times of expenditure restraint becomes all the more important, and underscores the need for upstream mechanisms to identify the projects that offer the best value for money and the greatest complementarity with other projects. Enhanced regional coordination between member states in the selection and building of infrastructure projects is an avenue to be explored in this context.
17. Scaling down public investment plans must be part of the policy adjustments in the near term to ensure sustainability. The pursuit of ambitious, pre-shock investment plans, at the time of a precipitous collapse in oil prices, is unsustainable. Fiscal conservation measures on expenditure should extend beyond the initial the scaling-up period, because oil revenue levels continue their downward trend. It should be noted that that even the final, more conservative reform scenario leaves enough fiscal space for significant public investment spending (more than 10 percent of GDP).
18. An upward adjustment of consumption tax rates (such as that of the value-added tax) could be considered. A small increase (e.g., half a percentage point) has a big impact, especially because it applies to the broader tax base that comes with higher growth per capita.
19. Finally, CEMAC authorities would benefit from securing alternative, less expensive sources of financing than borrowing at commercial rates. Although all scenarios assume that financing shortfalls would be met with external non-concessional debt, it should be understood that any effort to enhance the concessionality of financing would ease the burden of reform in other policy areas.
Buffie et al. (2012): “Public Investment, Growth and Debt Sustainability: Putting Together the Pieces” IMF Working Paper WP/12/144.
Melina et al. (2014): “Debt Sustainability, Public Investment, and Natural Resources in Developing Countries: the DIGNAR Model” IMF Working WP/Paper 14/50
IMF (2014) Staff Report on Common Policies for member Countries, Selected Issues PaperIMF Country Report 14/252.
Pritchett (1996): “Mind your P’s and Q’s: the cost of public investment is not the value of public capital” World Bank Policy Research Working (WPS 1660).
Prepared by Jean van Houtte, with assistance from Matteo Ghilardi.
IMF staff has developed a new Public Investment Management Assessment (PIMA) tool to this effect.
Pritchett (1996) estimates that public capital accumulation in less developed countries can be less than 50 percent of public investment; he also finds that “factor accumulation,”, including public capital formation, routinely exceeds 80 percent in other economies.