Turkey
Tenth Review Under the Stand-By Arrangement-Staff Report; and News Brief on the Executive Board Discussion

This paper assesses Turkey’s Tenth Review Under the Stand-By Arrangement (SBA). The IMF supports Turkey’s economic program under an SBA covering 2000–02. The arrangement was augmented with resources under the Supplemental Reserve Facility in December 2000, and under the credit tranches in May 2001. Since the ninth review, the authorities have kept macroeconomic policies in line with the strengthened program adopted in May 2001. All performance criteria for the tenth review have been met, as have those for the eleventh review.

Abstract

This paper assesses Turkey’s Tenth Review Under the Stand-By Arrangement (SBA). The IMF supports Turkey’s economic program under an SBA covering 2000–02. The arrangement was augmented with resources under the Supplemental Reserve Facility in December 2000, and under the credit tranches in May 2001. Since the ninth review, the authorities have kept macroeconomic policies in line with the strengthened program adopted in May 2001. All performance criteria for the tenth review have been met, as have those for the eleventh review.

I. Introduction

1. The Fund supports Turkey’s economic program under an SBA covering 2000-02 (Table 1). The arrangement was augmented with resources under the Supplemental Reserve Facility in December 2000, and under the credit tranches in May 2001. The ninth review under the program was completed on August 3, allowing a purchase in an amount equivalent to SDR 1.2 billion. The next purchase, in an amount equivalent to SDR 2.4 billion, is contingent on the completion of this review. Under the Fund’s safeguards assessment policy, the external audit assessment was completed on August 21, 2001 (Appendix I and Section IV). The World Bank supports Turkey under a country assistance strategy envisaging overall lending of up to US$6.2 billion during FY2001-03.

Table 1

Turkey: Schedule of Purchases Under the SBA/SRF, 1999-2002

article image

End-May performance criteria on consolidated government sector primary balance and central government primary expenditure.

End-August performance criteria on NDA and NIR; end-July performance criteria on consolidated government sector primary balance and central government primary expenditure, and end-June performance criteria on external debt.

End-October performance criteria on NDA and NIR; end-September for all other performance criteria.

2. In the attached letter, the authorities review developments and policies under their program, and request the modification and setting of several performance criteria and the completion of the tenth review. They request a modification of the ceiling for the performance criterion on the cumulative primary expenditure of the central government for end-December 2001, reflecting higher inflation than envisaged at the time of the ninth review. In addition, they request the setting of the performance criterion on the change in net international reserves for end-December to allow full carryover of the unused portion of the previous period’s limit. Such a move would permit the CBT to use the (delayed) tenth review disbursement of Fund resources to sterilize the on-lending of external resources to the Treasury. Moreover, to strengthen their disinflation effort, the authorities request the conversion of the end-December indicative ceiling for base money into a performance criterion. Finally, the authorities request that the net domestic assets of the Central Bank of Turkey (CBT), an indicative target for end-December, be measured using a fixed U.S. dollar/TL exchange rate.

3. In the letter, the authorities also indicate their intention to adopt a new medium-term economic program, for which they seek international financial support. Although the current program was showing signs of beginning to work from late August, the repercussions of the September 11 events on Turkey are likely to be severe, giving rise to additional external and fiscal financing needs. To ensure debt and balance of payments sustainability, any such financing will need to take place in the context of a new strengthened medium-term program, which the authorities plan to present to the Fund by end-2001.

II. Performance under the Program

4. Since the ninth review, the authorities have kept macroeconomic policies in line with the strengthened program adopted in May. All performance criteria for the tenth review have been met, as have those for the eleventh review (Annex A in the Attachment). Fiscal performance has been strong, with the authorities fully on track to meet this year’s ambitious public sector primary surplus target of 5½ percent of GNP. The CBT has kept monetary aggregates in line with program targets, although the moderation in base money growth also reflects increased demand for foreign currency deposits (Figure 1). Uncertainty over the exchange rate regime has abated since early August, following the CBT’s announcement that it would strictly limit its discretionary intervention. Consistent with the CBT’s mandate for price stability, interest rate decisions have made increasing reference to the need to contain inflation.

Figure 1.
Figure 1.

Turkey: Monetary Developments, 2001

(In trillions of TL)

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Source: Data from the Turkish authorities.1/ In millions of U.S. dollars.

5. The authorities have continued with structural reform, although steps to encourage foreign direct investment have been tentative, and there has been limited progress toward divesting large enterprises:

  • The program’s structural conditionality is being met. The end-October structural benchmark on appointing advisors for Turk Telekom was observed, and actions on two other benchmarks (submitting a Public Procurement Law to parliament and supporting the draft 2002 budget with accounts and financial outlook for the various public sector entities) were completed with only a short delay (Annex B in the Attachment).

  • Although the banking system remains under strain, there has been significant progress in implementing financial sector reform. State banks are being restructured, with the largest of them now among the most profitable in Turkey. The authorities have continued to press private banks into achieving and maintaining an 8 percent capital adequacy requirement, and to resolve the remaining intervened banks. However, the combination of economic recession, exchange rate instability, and high interest rates has placed the banking system under strain, despite these considerable reform efforts.

  • The agenda for encouraging foreign direct investment is gradually taking shape. A mid-September conference on administrative barriers to investment formulated an action plan, which was submitted to the Council of Ministers in mid-November. A high level investor conference has, however, been postponed into next year.

  • Privatization is facing delays. While the conditionality on Türk Telekom was met, legislative and administrative delays make it unlikely that the privatization plans for TEKEL (tobacco and alcohol monopoly) and ŞEKER (sugar company) will be ready by year-end. Moreover, the planned sales of shares in TÜPRAŞ (oil refinery) and POAŞ (petrol distribution) will not take place this year because of weak market conditions in the aftermath of September 11.

6. After some initial problems with program implementation, the long-awaited decline in interest rates started toward the end of the summer (Figures 2-3). The May program was designed to break a crisis of confidence, helping interest rates to decline and concerns about debt sustainability and rollover to diminish. However, amid delays in meeting the conditions for completing the ninth review, political disagreements over program implementation, and the lack of a nominal anchor, interest rates rose to more than 100 percent in July, adding to concerns over debt sustainability. But by late summer—after a period relatively free from political tension, the floating exchange rate being given a chance to work, and the ninth review completed (and the tenth review seemingly on track)—the benchmark bond rate had fallen to 80 percent, and the exchange rate was broadly stable. In addition, the voluntary debt swap in June, the measures introduced in July to encourage investments in government paper (including withholding tax reductions, partial remuneration of reserves, and the possibility of increased state bank participation) and increasing use of FX-linked borrowing all helped to ease the debt rollover problem for the remainder of 2001.

Figure 2.
Figure 2.

Turkey: Market Developments, 2001

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Source: Data from the Turkish authorities.
Figure 3.
Figure 3.

Turkey: Interest Rates, 2001 1/

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Source: Data provided by the authorities.1/ Compounded interest rates.

7. Just as the program was beginning to show signs of working, the adverse economic effects of September 11 set back hopes of a rapid recovery. Within days interest rates on the benchmark bond rate shot up by 15 percentage points, the stock market lost almost 20 percent of its value, and the exchange rate fell by 10 percent. In part this reflected investors’ increased aversion to emerging market assets, in Turkey’s case compounded by its vulnerability (in terms of indebtedness) and its location. This sizeable external shock is affecting the economy through many channels: lower partner country demand, loss of tourism receipts, reduced access to international financial markets, and weakened privatization and FDI prospects. All of these have increased the risk to the external position and fiscal sustainability (Table 2 and Appendix II). At the same time, the more depreciated exchange rate has also contributed to the recent rise in inflation (to over 66 percent year on year in October for CPI), despite the authorities’ adherence to the monetary program.

Table 2

Turkey: Indicators of External Vulnerability, 1997-2003 1/

article image
Sources: Data provided by the Turkish authorities; and Fund staff estimates and projections.

For 2001-03, program projections.

As of end-June 2001, reserves stood at US$ 17.5 billion.

By residual maturity.

Interest plus medium-and long-term debt repayments as percent of current account receipts (excl. off. transfers).

As of August 2001.

As of June 2001.

As of August 2001.

Deflated by the WPI.

For 2001, as of November 19.

8. While confidence has returned in recent weeks, Turkey’s reform program remains precariously, if auspiciously, poised. While the challenges, most notably ensuring the sustainability and rollover of the government’s domestic debt, remain serious, progress was being made pre September 11. Following weaker-than-projected growth in the first half of the year, there were signs that the economy was bottoming out (Figures 4-5). Export growth has also been strong, despite slowing world growth. And despite the immediate impact of the September 11 shock, in the last month market conditions have improved considerably, largely on the hope of international financial assistance from the official sector. In recent weeks, benchmark bond rates have fallen to around 75 percent, the exchange rate has appreciated somewhat, and the stock market has recovered strongly. Also, the spreads on Turkish bonds in the Euromarket have narrowed, despite developments in Argentina (Figure 6). Thus, while many challenges remain, the prospect of a virtuous spiral of improved confidence in rolling over the debt, lower interest rates, and exchange rate strengthening might once again take hold.

Figure 4.
Figure 4.

Turkey: Output and Demand, 1992-2001

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Source: Data provided by the Turkish authorities.1/ The stock of loans has been adjusted for the shift to nonperforming loans of the loans of the banks taken over in December 1999 by the Saving Deposit Insurance Fund.2/ Seasonally adjusted; the total VAT has been adjusted for tax changes.
Figure 5.
Figure 5.

Turkey: Inflation, 1996-2001

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Source: Data provided by the Turkish authorities
Figure 6.
Figure 6.

Turkey: Bond Spreads, 2001

(percentage Points)

Citation: IMF Staff Country Reports 2002, 021; 10.5089/9781451838039.002.A001

Data provided by the authorities; and Bloomberg.

III. Report on the Discussions

9. Initially straightforward discussions for the tenth review had to be adapted following the events of September 11. During the September 6-19 discussions, the staff confirmed that all performance criteria relevant for the review had been met, but understandings could not be reached on an ambitious inflation target for 2002, or on measures to reach the 2002 fiscal targets. On top of this, the events of September 11 argued for a shift in focus. Subsequent discussions in Washington and Ankara covered not only the measures required to conclude the tenth review but also the additional financing and policy requirements in light of changed circumstances. This section reports on discussions on all these topics.

A. Macroeconomic Outlook and Financing Needs

10. Reflecting developments through the third quarter, including the events of September 11, the authorities and staff agreed to revise the macroeconomic framework8 and Table 3):1

Table 3

Turkey: Selected Indicators, 2000-03

article image
Sources: Data provided by Turkish authorities; and Fund staff estimates.

Average of monthly nominal interest rate divided by 12-month ahead CPI inflation.

On a commitment basis, excluding profit transfers from the CBT, interest receipts, and privatization proceeds.

Interest payments minus interest receipts plus profit transfers from the central bank, including receipts from revaluation accounts.

Interest payments minus interest receipts plus CBT profits before transfers to the government.

Change in reserve money in percent of GNP, where reserve money is defined as currency issued plus reserve requirements.

For 2001 these include the privatization proceeds from the Privatization Administration’s Portfolio.

  • The delayed impact of the May program and the events of September 11 justified downward revisions to GNP projections for this year and next. The decline in GNP was revised to 8½ percent for 2001, compared with the 5½ percent decline envisaged at the time of the ninth review. Even before September 11, a steeper-than-expected decline in GNP in the second quarter and weaker-than-envisaged prospects for agricultural production had suggested that economic recovery would be delayed. The worsened external environment in the aftermath of the events of September 11 compounded this delay. It was expected, however, that as the program took hold, the economy would rebound toward the end of the year, with real GNP growth of 4 percent forecast for 2002, mainly as a result of stock building and a modest increase in private consumption.

  • The inflation targets for 2001-02 were raised. While seasonally adjusted monthly inflation had declined to 3 percent in July-August, it rose to over 4 percent in September-October, in large part owing to faster-than-projected currency depreciation after September 11. With little scope for a substantial tightening of monetary policy in light of debt rollover concerns, and thus the risk that currency depreciation might ensue, it was the joint assessment of the staff and authorities that CPI inflation would likely reach 65 percent by year-end, up from 58 percent projected at the time of the ninth review. As for 2002, the staff initially argued for a target closer to the 20 percent envisaged in the May program but, given the events of September 11, the deeper than expected recession, and the continued concerns about domestic debt rollover, in the end accepted the authorities’ 35 percent target.

  • In light of the increased uncertainty created by the events on September 11, the macroeconomic framework was, however, seen to be subject to risks. In the staffs view, the risks for growth in 2001-02 were mainly on the downside, while inflation in 2001 could end up above the present projections. While acknowledging the risks, the authorities argued that revision of the macroeconomic framework should await forthcoming data (such as figures for GNP in the third quarter and inflation in November which will be out by early December) for confirmation of these risks. The authorities indicated their readiness to take additional measures as needed to reach the fiscal targets should the macroeconomic framework change.

11. As regards the balance of payments, the authorities and staff estimated that, following September 11, Turkey was likely to face an external financing gap of about US$10 billion in 2001-02. (This gap and its sources are discussed in detail in Appendix III; the detailed balance of payments projections are shown in Table 4). Both sides stressed that these projections were subject to much uncertainty, and that employing a range of US$8-12 billion was, therefore, appropriate. On the current account, it was a common view that travel receipts would be particularly adversely affected (with a projected decline of US$2 billion in receipts next year), given Turkey’s location. On the capital account, despite the recent € 500 million Eurobond issue, the authorities indicated that issuance would be constrained in the post-September 11 environment to about US$2 billion next year (compared with a pre-September 11 plan for issuance of US$3.5 billion or more). And, while net banking sector outflows were expected to moderate significantly compared with 2001, it was agreed that net outflows of about US$5 billion next year were likely unless international market confidence was restored quickly, or the program contained a strong external private sector involvement (PSI) element. Regarding the latter, it was noted that the experience with PSI had so far been disappointing—despite meetings with international banks in December 2000 and June 2001 on maintaining their exposures to Turkey, outflows of some US$7 billion had been recorded under interbank credit lines during 2001 so far—and that successful external PSI would only occur if explicitly supported by the appropriate authorities in those countries whose banks are the major players.

Table 4

Turkey: Balance of Payments, 1998-2003

(In billions of U.S. dollars)

article image
article image
Sources: Data provided by the Turkish authorities; and Fund staff estimates and projections.

The decline in other receipts between 1998 and 2000 partly reflects a methodological change in the compilation of this item.

Including privatization receipts.

Nonbank external debt less the NFA of the banking system.

Interest plus medium-and long-term debt repayments as percent of current account receipts (excluding official transfers).

12. Although rollover prospects through end-2001 looked reasonable, the September 11 shock made the task of rolling over domestic debt more challenging next year:

  • The staff commended the authorities for their flexible management of domestic debt during August and early September, which put the Treasury in a comfortable cash position and helped to cushion the impact of the shock on the public finances. The larger-than-programmed primary surpluses and the measures introduced in July to increase the demand for government paper had ensured good prospects for a smooth rollover for the remainder of 2001 (¶10). In managing the rollover in 2002, the authorities stressed that they would continue to respond flexibly to changes in market conditions, exploring ways of lengthening maturities while meeting the balance sheet concerns of investors through instruments such as fx-linked securities and floating rate notes (¶11). They had also established an interagency committee to assist the Treasury in the formulation of its borrowing policy (¶12).

  • The shock had worsened budget financing prospects for 2002, with well over half of the additional external financing in 2002 being needed for the budget to support a smooth rollover of domestic debt. Financial market conditions abroad had worsened prospects for Eurobond issues and privatization proceeds, which in the absence of additional external financing would increase the financing to be raised domestically—mainly from the banking sector. While stable deposit growth and the strong financial position of state banks should allow them to contribute significantly to financing the budget, the remaining financing needed from private banks would be well in excess of what they could absorb without unduly crowding out private sector credit growth (Table 5). While this risk would very much depend on the quality of policies and the circumstances of the moment, projected balance sheet developments for public and private banks suggest that over half of the additional external financing would need to go to the budget to ensure that the private sector rollover rate was held to credible levels (around 85 percent). A full analysis will be provided in the documentation supporting the new stand-by arrangement.

Table 5

Turkey; Government Borrowing and Bank Capacity, 2001-02

article image

Assuming private sector lending grows in line with nominal activity and, consistent with the balance of payments, external interbank borrowing declines by US$3 1/2 billion.

Assumes higher than anticipated nonperforming loans and lower deposit growth.

For comparability, includes valuation changes in the stock of debt.

13. In the staffs assessment, with additional external financing of US$10 billion in 2001-02 and continuation of strong policies, Turkey’s debt and balance of payments could be sustainable over the medium term:

  • Provided the additional US$10 billion is of medium-or long-term maturity and confidence in Turkey’s assets returns, the external position should be sustainable for the next several years—that is, current account positions should be financed with identified inflows while reserves remain at safe levels (Appendix III, Tables 15 and 16). Medium-and long-term financing would help lower the ratio of short-term debt (by remaining maturity) to reserves—widely acknowledged to be the most helpful single external reserves indicator—to less than 150 percent through end-2004. Although under conservative assumptions the ratio would begin to pick up in 2005, reserves would likely turn out to be much higher in the medium term than currently projected if the program succeeds in restoring market confidence as intended. The current account is projected to record moderate deficits of about US$2 billion (1 percent of GNP) over the medium term. These projections imply broadly unchanged shares of exports, imports, and services in GNP from 2002 onward. At the same time, the capital account balance is projected to move from a US$3 billion deficit in 2002 to a surplus of US$3-6 billion over the medium term, reflecting an improvement in confidence. More than half of the turnaround reflects the assumption that banks will begin to secure modest net capital inflows, with the remaining improvement accounted for, in declining order of importance, by better international capital market access by the Treasury, a turnaround in net international corporate sector lending, and a gradual improvement in net portfolio flows.

  • With over half of the additional external financing going to the budget, Turkey’s public debt would remain sustainable in the medium term—that is, it would shift to a declining trend relative to GNP—provided real interest rates converge to program levels and the primary surplus remains strong (Table 6). The sharp increase in the public sector’s net debt-to-GNP ratio expected for 2001 (to some 95 percent from last year’s 57 percent) mostly reflects the one-off factors of the costs of bank recapitalization, the fall in output, and the real depreciation of the exchange rate. With US$5-7 billion in additional external financing, and with conservative (high) estimates of real interest rates, the public sector debt-to-GNP ratio is expected to fall by about 10 percentage points in 2002. Assuming a continued strong public sector primary surplus, a recovery of growth to 5 percent, and a small real appreciation, the debt-to-GNP ratio is projected to fall further to about 75 percent in 2003 and continue falling over the medium term.

Table 6

Turkey: Medium-Term Public Debt Sustainability, 2001-06 1/

(In percent of GNP)

article image

Assuming US$10 billion additional external financing in 2002, of which US$7 billion for budget financing.

For projections, defined as an ex ante real interest rate using programmed CPI inflation, based on an average maturity of six months.

B. Fiscal Policy

14. The authorities were fully on track to meet their 2001 public sector primary surplus target of 5.5 percent of GNP (¶16 and Tables 7-9). In the authorities’ assessment, revenue overperformance and delays in investment expenditure in the broader public sector would more than offset higher spending in the central government generated by indexed wages and pensions. The authorities also insisted that expenditure arrears were being minimized by exerting tight control, and by handling any excess commitments, revealed via regular surveys of key ministries, through the contingency. The staff expressed concern that a properly formulated and comprehensive arrears survey be undertaken as soon as possible. The staff nonetheless recognized the strong overall budget performance, and argued that the actual outturn could be even stronger, due to conservatively forecast revenues. If this was the case, the authorities agreed to save any overperformance in central government revenues given the government’s financing needs. The authorities requested a modification of the end-December performance criterion on primary expenditure, to reflect the impact of higher than programmed inflation on indexed wages and pensions. In view of the likely overperformance on the primary balance, and since expenditures for the year will still decline by 4½ percent in real terms, the staff supports the authorities’ request.

Table 7

Turkey: Central Government Primary Budget, 2000-02

(In percent of GNP)

article image
Sources: Data provided by Turkish authorities; and Fund staff estimates.

Excluding privatization proceeds.

Excluding recapitalization of state banks.

Excluding privatization proceeds, CBT profits, and interest receipts.

Table 8

Turkey: Central Government Primary Budget, 2000-02

(In trillions of Turkish liras)

article image
Sources: Data provided by Turkish authorities; and Fund staff estimates.

Excluding privatization proceeds.

Excluding recapitalization of state banks (for 2001 only).

Excluding privatization proceeds, CBT profits, and interest receipts.

Table 9

Turkey: Public Sector Primary Balances, 2000-02

article image
Sources: Turkish authorities; and staff estimates.

Excluding privatization proceeds, transfers from the CBT, and interest receipts.

Excluding recapitalization of state banks.

Excluding severance payments for retirees amounting to TL 201 trillion (0.1 percent of GNP).

Added to the public sector balance for 2002. In 2001 they are expected to show a TL 40 trillion primary surplus.