Selected Issues

This paper reviews the progress made by the Tunisian authorities in their efforts to liberalize the capital account and highlights the potential benefits of implementing the remaining reforms in this area. It also analyzes the developments within a unified analytical framework, provides tentative insights regarding priorities for the government’s growth strategy ahead, and discusses the impact of ongoing labor market reforms and investment promotion policies.


This paper reviews the progress made by the Tunisian authorities in their efforts to liberalize the capital account and highlights the potential benefits of implementing the remaining reforms in this area. It also analyzes the developments within a unified analytical framework, provides tentative insights regarding priorities for the government’s growth strategy ahead, and discusses the impact of ongoing labor market reforms and investment promotion policies.

I. Tunisia : Potential Growth-Enhancing Effects of Further Capital Account Liberalization and Supporting Reforms1

A. Introduction

1. Capital account liberalization is a key element of the Tunisian authorities’ strategy for reaching emerging market OECD countries income levels and reducing unemployment. The authorities are gradually opening the capital account in order to tap external savings, diversify balance of payments financing, allow for portfolio diversification, and improve the efficiency of domestic financial markets.

2. The purpose of this paper is to illustrate progress made by the Tunisian authorities in their efforts to liberalize the capital account and highlight the potential benefits of implementing the remaining reforms in this area. Key reforms have been achieved in support of capital account liberalization: (i) the authorities have established a track record of sound macroeconomic policies; (ii) the prudential framework for the financial sector has been strengthened; (iii) a modern legal framework has been established; (iv) the infrastructure for the capital market has been developed; (v) some restrictions on the holding of foreign exchange by residents, and on foreign investment have been removed and conditions on external borrowing both by banks and firms have been relaxed; (vi) the trade regime has been significantly liberalized; and (vii) significant progress has been made toward the strengthening of the systemic liquidity framework.

3. Despite progress achieved in the area of capital account liberalization, there remains sizeable room for further reforms to maximize the potential benefits of liberalization. There are many opportunities for further reform. In particular, the strengthening of the banking system and the systemic liquidity framework, as well as a reduction of the level of external public debt are required before the implementation of most of the remaining agenda for capital account liberalization.

4. Recent evidence suggests that capital account liberalization does not only allow a diversification of the balance of payments financing sources, but could also contribute to boosting foreign direct investment by accelerating the development of financial markets. The implementation of supportive measures to move forward with further capital account liberalization are part of the authorities’ objective to develop domestic financial markets and attract foreign investment to boost growth.

B. Capital Account Liberalization and Supporting Reforms : Progress Achieved and the Remaining Steps

The capital account liberalization process: description and progress achieved

5. The Tunisian authorities have developed, in consultation with staff, a three-phase plan to liberalize the capital account2:

  • The first phase consists of reforms aimed at liberalizing medium to long-term flows such as nonresident direct investments and long-term loans to listed firms, limited nonresident investments in local currency government securities, and other measures to enhance the overall effectiveness of financial intermediation and diversify the balance of payments financing sources.

  • The second phase involves liberalizing direct investment by Tunisians abroad, allowing overseas portfolio investments by institutional investors and portfolio investments by nonresidents in debt instruments. This stage includes a transition to a floating exchange rate, a deepening of the foreign exchange market, and a banking system that is sufficiently strong to withstand international competition. Progress would also be needed in the development of government securities markets to increase market liquidity.

  • The third phase involves full currency convertibility. It entails liberalizing domestic portfolio investment abroad and loans by residents to nonresidents. Moving to this stage would require a robust financial sector and a resilient balance of payments position.

6. The main measures of this capital account liberalization strategy, as well as the status of their implementation are summarized in Table 1.

Table 1.

Sequence of Capital Account Liberalization

article image
article image

The level of priority is for the whole process of capital account liberalization. Measures with smaller numbers have a higher priority.

7. The first phase of the capital account liberalization process in Tunisia is almost complete. The surrender requirement of foreign exchange proceeds has been eliminated. Regulations on foreign exchange holdings in Tunisian banks have been relaxed. Restrictions on inward nonfinancial foreign direct investment have been lifted. Foreigners may invest freely in most economic sectors. The approval of the High Investment Commission is no longer required for the acquisition by foreign nationals of Tunisian securities entailing voting rights. The authorities have allowed nonresident investment in local-currency government securities, subject to a ceiling of 10 percent of the outstanding stock. Resident financial institutions can now contract foreign currency loans from nonresidents in maturities of over 12 months without limits (previously, TD 10 millions a year). Rated resident companies can contract such loans up to an annual limit of the equivalent of TD 10 millions (previously TD 3 millions a year). For short-term foreign currency loans, borrowing limits have been significantly increased for both financial institutions and corporations. The liberalization of external borrowing was accompanied by a relaxation of the regulation on foreign exchange lending to corporations.

The capital account liberalization process: the remaining steps

8. In order to further deepen the foreign exchange market, which is an important requirement of the capital account liberalization process, the authorities could speed up the abolishment of the nivellement policy by which banks must transfer all their correspondent account foreign exchange holdings to the BCT at the end of the day. While this policy allows the BCT to manage the foreign exchange reserves of the private sector by placing them on international markets, it is not conducive to building the corresponding skills at the commercial banks. Eliminating this requirement would enhance competition for exporters’ deposits as banks will be able to manage these deposits more effectively. As banks gain experience in managing foreign exchange assets, they could, in turn transfer this experience to the local foreign exchange market. The BCT could continue to closely monitor banks’ foreign exchange operations to ensure that there is no unauthorized capital movements.

9. Moving to the second phase of the authorities’ capital account liberalization plan will also require :

  • strengthening the banking system. The recent Financial System Assessment Program (FSAP) update mission indicated that the high level of non-performing loans (NPLs) remained the key vulnerability of the Tunisian banking system, although they do not constitute a systemic risk. Other factors contributing to its weakness include a lack of reliable financial data on enterprises, weaknesses in loan recovery, especially in the realization of collateral, and the relatively limited role of asset management companies.

  • addressing the remaining shortcomings in the systemic liquidity framework. On the money market, daily average turnover remains low and interest rates mirror those on BCT operations, with limited volatility within a narrow corridor. On the securities market, the Treasury bill holdings in the BCT portfolio have been modest, thus limiting the volume of open market operations. Furthermore, the maturity range of government securities has not been wide enough to help build a full yield curve. On the foreign exchange market, restrictions remain on forward operations and foreign exchange options. These shortcomings hamper further progress toward an open capital account as they constitute obstacles to capital markets development. Indeed, if a full opening of the capital account were to lead to large capital flows, the lack of preparedness of commercial banks, or more generally participants in the money and foreign exchange markets, which would be called to intermediate these flows could exacerbate the vulnerabilities in the financial system in the event of a sudden reversal.

  • reducing the level of external debt. If capital flows were to be fully liberalized, the amount of foreign exchange needed to service the current level of the external debt (68 percent of GDP at end 2005) could put significant pressures on the foreign exchange market, and could affect the credibility of the current exchange rate regime.

C. The Potential Growth Benefits of Further Capital account Liberalization and Supporting Reforms

10. It is difficult to isolate the growth impact of reforms aimed at capital account liberalization. In theory, there are reasons to expect capital account liberalization to have a positive impact on a country’s growth performance. Capital mobility lowers the cost of capital, reduces the cost of transferring technology and management know-how to the host country, and facilitates greater competition and the development of domestic financial markets. Moreover, liberalization of capital flows typically boost foreign direct investment, which enhances growth prospects. It facilitates consumption smoothing by allowing consumers to borrow when their incomes fall and repay when their income rises, thus promoting longer term economic stability. It also leads to portfolio diversification, which helps the development of financial markets and allows for risk-sharing among investors, thus contributing to financial stability. Despite these theoretical benefits, empirical evidence on the general impact of capital account liberalization on economic growth is inconclusive. Part of the reason for this is the fact that capital account liberalization is often an element of a more broad-based reform package. In addition, difficulties in defining appropriate measures of liberalization reduce the scope of any analysis aimed at isolating the direct or indirect effects of capital account liberalization on growth 3. Another important reason why empirical studies have generally failed to detect a strong positive correlation between capital account liberalization and growth is that many countries liberalized their capital account when they were not ready 4. Even when countries are ready, capital account liberalization carries risks which could destabilize financial systems if regulatory frameworks are not adequate and if supervision is weak.

11. Despite difficulties in finding appropriate measures of the degree of capital account liberalization, the magnitude of portfolio investment flows could give a fairly good idea of a country’s remaining restrictions on capital flows. Given the fact that most countries have liberalized most foreign direct investment flows, the size of portfolio investment flows is a good indicator of a country’s remaining restrictions on capital flows. In Table 2 and Table 3, portfolio flows to and from Tunisia are relatively low, consistent with the fact that capital flows to Tunisia remain relatively restricted. For example, in 2004 and during the period 1995-2004, portfolio flows were on average less than 0.1 percent of GDP in Tunisia.

Table 2.

Selected Countries’ Balance of Payment Flows, 2004

article image
Source: Tunisian authorities and IMF staff estimates.
Table 3.

Selected Countries’ Balance of Payment Flows, 1995-2004 average

article image
Source: Tunisian authorities and IMF staff estimates.

12. Recent experiences from selected countries suggest that capital account liberalization has helped ease the constraints on the financing of balance of payments (BOP), and therefore supported investment. An analysis of selected countries’ balance of payments flows shows that portfolio investment flows can play an important complementary role to the financing of a country’s current account deficit. In 2004 for example, the contribution of portfolio investment liabilities flows to the financing of the current account deficit has been close to or above the contribution of foreign direct investment in the Czech Republic, Hungary, Mexico, Poland, Portugal, and Turkey (Table 2). Portfolio flows have clearly provided some flexibility to these countries in the financing of their trade and current account deficits. It is worth mentioning however that foreign direct investment (FDI) has the advantage of not augmenting external debt and tends to be more stable. This point is particularly important for Tunisia, which already has a high level of external debt.

13. Recent experience also suggests that the degree of openness to portfolio flows tends to be positively correlated with FDI. In order to capture a country’s openness to portfolio flows, one can consider the magnitude of the sum of the absolute values of portfolio assets and liabilities flows. A country is more open to portfolio flows when it allows relatively large amounts of portfolio flows in either direction. The indicator of openness to portfolio investment used is based on the ratio of this sum to the country’s GDP to adjust for the size of the economy 5. Using this indicator, recent experience suggests that openness to portfolio investment tends to be positively correlated to the level of foreign direct investment (Chart 1).

Chart 1.
Chart 1.

FDI and Portfolio Investments (In percent of GDP, average 2000-2004 or latest year available)

Citation: IMF Staff Country Reports 2006, 208; 10.5089/9781451837889.002.A001

Source: IMF staff estimates.

14. Although the positive correlation between the degree of openness to portfolio flows and FDI does not necessarily imply a causal relationship, it supports the view that capital account liberalization could boost FDI by accelerating the development of financial markets, and so enhancing the attractiveness of an economy. The positive correlation between the degree of openness to portfolio flows and foreign direct investment that is illustrated in Chart 1 could also reflect the fact that countries that open portfolio flows tend to have relatively developed financial markets. These markets could offer various flexible sources of financing to investors, thus providing them with the necessary incentives to invest their capital in these countries, including in the form of equity. However, a deeper analysis is needed to fully understand the relationship between the degree of openness to portfolio flows and FDI.

D. Conclusion

15. The authorities are implementing a broad-based reform agenda in which increasing openness to the rest of the world economy is an important pillar, and evidence supports the view that capital account liberalization will help Tunisia reach its growth objectives. Significant progress has been made toward an open capital account. However, additional reforms are needed to further liberalize the capital account, deepen financial markets and boost investment. Increased momentum in the implementation of these reforms will be essential to bringing Tunisia to a higher-growth path.


  • Eichengreen, Barry, 2001, “Capital Controls: Capital Idea or Capital Folly, “ Policy Options, Institute for Research on Public Policy, July-August.

    • Search Google Scholar
    • Export Citation
  • Edison, Hali J., Michael W. Klein, Luca Antonio Ricci, and Torsten Sløk, 2004, “Capital Account Liberalization and Economic Performance: Survey and Synthesis.IMF Staff Papers, International Monetary Fund, Vol. 51 (2), pp.220 -256.

    • Search Google Scholar
    • Export Citation
  • Forbes, Kristin J., 2004, “Capital controls: Mud in the Wheels of Market Discipline,NBER Working Paper 10284 (Cambridge: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2005, “The Economic Impact of Controls on Capital Movements: Preliminary Considerations”, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2002, “Capital Account Liberalization and Financial Sector Stability”, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Laurens, Bernard and Abdourahmane Sarr, 2002, “Liberalization of the Capital Account in Tunisia—Progress Achieved and Prospects for Full Convertibility.”, SM02/120 Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

Table A1.

Sequence of Capital Account Liberalization

article image

Prepared by Jacques Bouhga-Hagbe.


For a more detailed discussion, see Laurens and Sarr (2002).


For further discussion on countries’ experiences in the sequencing and coordination of capital account liberalization with other policies, see International Monetary Fund 2002.


One should note that this indicator of a country’s restrictiveness to capital flows has significant drawbacks. When one adjusts for the size of the economy, some countries such as the US that are open to portfolio flows end up with a value of the indicator that is relatively low because their GDP is very large compared to the size of their portfolio flows. However, if one does not adjust for the size of the economy, the indicator used could be misleading as larger economies are likely to have larger volumes of portfolio flows without necessarily having fewer capital account restrictions. The indicator used has the advantage of being relatively simple, and is considered only for illustrative purposes.

D. Appendix: Description of the Model

28. In a perfect foresight environment, consumers choose consumption, labor input, and investment to maximize utility Σt=1βt[logct+Ψlog(llt)]Nt

subject to ct+(1+τkt)[(1+η)kt+1(1δ)kt]+xt(1τlt)wtlt+rtktTt,

where, c is consumption, Ī a maximum allocation of time for work (set to 5000 hours per year), l hours worked, k the beginning-period capital stock, x net exports (all in per-capita terms), N the population, 1/(1+τk) the investment wedge, (1-τl) the labor wedge, T lump sum taxes, w the wage, r the rental rate for capital, β the discount factor, η the population growth rate (set to 1.5 percent), and δ the depreciation rate (5 percent). Note that the terms in square brackets in the budget constraint is per-capita investment.

29. The labor and investment wedges represent any distortion in the labor market or investment environment, respectively. The former resembles a tax on labor income, while the latter is represented as an additional cost of investment (as in Chari, Kehoe, and McGrattan, 2004), which can be thought of as administrative costs of undertaking an investment. However, this interpretation of the investment wedge should not be taken literally. In fact, specifying the investment wedge as an implicit tax on capital income yields very similar results.

30. Firms’ production function is yt=Atka((1+γ)tl))1a, where γ represents a labor productivity growth trend. Firms choose labor and capital in order to maximize profits Atkta((1+γ)tlt)1artktwtlt in each period.

31. All variables (except labor) are detrended by the labor productivity trend, assumed to be 2 percent per year, corresponding roughly to productivity growth in the more dynamic OECD countries. Hence, an increase (decline) in a detrended variable can be interpreted as Tunisia converging toward (diverging from) emerging OECD economies. Detrended per capita variables are written and defined as Zt̂=Zt(1+γ)t.

32. Equilibrium is defined by consumers’ and firms’ first-order conditions, the production function, and a resource constraints as follows:


where ĝ is detrended per-capita government spending. The sum of ĝ and can be thought of as an ‘income accounting’ wedge between GDP on one hand and consumptions and investment on the other.


33. In this deterministic version of the model, the labor, investment, and productivity wedges can be derived directly from equations (1), (2), and (5), respectively. The income accounting wedge is taken directly from the data. The model is calibrated such that 1999 corresponds to the steady state. In this regard, the notion of steady state should not be taken literally but rather be thought of as a benchmark, convenient for distinguishing the impact of the various wedges. The labor and investment wedges are normalized to 1 (i.e. τl and τk are zero) in the steady state. Moreover, since all variables are constant in the steady state, equations (1) and (2) become (with SS indicating steady state):


34. Furthermore, the law of motion k^t+1=(1δ)k̂t+ît(1+η)(1+γ) yields the steady state capital GDP ratio, where î is detrended per-capita investment:


35. These three steady state conditions, together with consumption, GDP and labor data from 1999 are used to calibrate the values of the discount factor (β), and the consumption-leisure parameter (Ψ). Actual hours worked are not available; instead it is assumed that all employed workers work 40 hours per week. As a consequence, changes in labor input are seen only as changes in employment and misses variations in hours per worker. This is likely to be a reasonable first approximation, however. The initial (1999) capital stock is set to satisfy the steady state capital/GDP ratio, and subsequently follows the law of motion above. Accordingly, the calibration yields a β of 0.94, a Ψ of 9.4, and a steady state capital-GDP ratio of 2.4.


Prepared by Ludvig Söderling.


Due to a recent change in methodology, to follow ILO standards, there is a break in the unemployment series in 1999, complicating comparisons of levels before and after 1999. It is clear, however, that the declining trend in unemployment began around 1999.


Chari, V., P. Kehoe, and E. McGrattan (2004), “Business Cycle Accounting”, Federal Reserves Bank of Minneapolis Staff Report No. 328.


The model also contains an ‘income accounting’ wedge (see appendix), but this wedge turns out to be relatively unimportant quantitatively and will thus not be discussed further.


The partial reversal of both wedges in 2004 may simply be an issue of the data, which are still preliminary. What is important, however, is the general tendency, rather than one particular data point.