Journal Issue


International Monetary Fund
Published Date:
February 2000
  • ShareShare
Show Summary Details
Turkey: Basic Data
Area:774,815 square kilometers
Population 1998:63.5 million
Labor force 1998:22.5 million
GNP per capita 1998:US$3,051
19951996199719981999 1/
Real economy(Percentage change)
Real GNP7.
Domestic demand12.
CPI (end-of-period)78.979.899.169.764.7 5/
Unemployment rate (in percent)
Gross national savings (in percent of GNP)21.918.920.823.220.0
Gross domestic investment (in percent of GNP)25.223.424.623.922.8
Public finance(In percent of GNP)
Consolidated budget balance-4.0-8.2-7.6-7.1-12.1
Public sector borrowing requirement4.813.113.0151823.3
Net debt of the public sector41.346.542.944.558.0
Money and credit(End-year, percentage change)
Broad liquidity 2/106.4113.1117.676.082.6
Reserve money 3/79.291.5100.880.380.5
Credit to private sector131.7120.5125.881.758.3
Interest rates(Year average)
T-bill rate 4/125.8132.4105.2115.7107.7 5/
Overnight money market rate108.2115.8101.4111.9109.9 5/
Balance of payments(In percent of GNP)
Trade balance-5.1-5.7-8.0-7.0-5.8
Current account balance 6/-0.5-1.3-1.40.9-0.5
External debt43.143.247.550.454.7
Of which: short term9.211.111.713.415.4
Reserves (US$ billion, end-of-period)13,81217,69519,57520,11224,395 7/
Fund position (as of October 31, 1999)
Holdings of currency (in percent of quota)135.5
Holdings of SDRs (in millions of SDRs)1.7
Quota (in millions of SDRs)964.0
Exchange rate
Exchange rate regimeFloating exchange rate
Rate on November 30, 1999TL 512,738 per US$
Real effective rate (1990=100)85.787.793.4101.3108.5 7/
Sources: Information provided by the Turkish authorities; and Fund staff estimates.

Staff projections.

Includes foreign currency deposits and repos.

Including banks’ foreign currency deposits at the central bank.

Simple average across maturities ranging from three months to one year, net of tax.


Central bank current account data differ from the national income accounts.

As of September 1999.

Sources: Information provided by the Turkish authorities; and Fund staff estimates.

Staff projections.

Includes foreign currency deposits and repos.

Including banks’ foreign currency deposits at the central bank.

Simple average across maturities ranging from three months to one year, net of tax.


Central bank current account data differ from the national income accounts.

As of September 1999.

I. Inflation as a Fiscal Problem1

1. This chapter revisits an age-old question: what is the link between budget deficits and inflation? The issue is particularly relevant for Turkey which, over the past three decades, has suffered from high and accelerating rates of inflation, without ever degenerating into full-blown hyperinflation. During the 1970s, inflation averaged about 20 percent, rising to 40 percent during 1981–87, 65 percent during 1989–93, and 85 percent during 1995–97. In late 1997, inflation reached 100 percent, before the government launched its most recent attempt at disinflation, in early 1998.

2. Over the past few years, the government’s Large financing requirement (including not only the central government, but also the extrabudgetary funds, the local authorities, the social security institutions, and the financial and nonfinancial state economic enterprises) would have resulted in an exploding debt-GDP ratio were it not for the monetization of these deficits through increasingly higher inflation rates. Large budget deficits are, in turn, the result of both substantially negative primary budget balances and high and rising real interest rates. In a chronic inflationary environment deprived of a monetary anchor, high real interest rates becomes in fact both the cause and the consequence of high inflation, that is, they feed into high inflation and in turn are fed by high inflation and the associated risks. If chronic and high inflation has not degenerated into hyperinflation as it did in most other countries, this is to be credited, at least in part, to the very slow pace at which inflation has eroded the demand for base money—a trait unique to Turkey.

3. Within Turkey there is often the perception that inflation is largely an issue of inertia—and that if inflationary expectations could somehow be broken, the problem would be solved. Undoubtedly, inflationary expectations are now firmly entrenched and any viable disinflation strategy must take them into account. But the fact remains that the weak fiscal position is the underlying cause of inflation in Turkey, and solving the fiscal fundamentals is the first necessary step in achieving disinflation; this is the focus of this chapter.

4. The chapter is organized as follows. Section A sets the stage for the discussion of the link between inflation and public sector deficits by reviewing the evolution of the primary balance of Turkey’s public sector, its net debt, real interest rates, and the holdings of base money during the 1990s, since—as explained in Section B—the crucial elements in the nexus inflation/debt dynamic are the primary fiscal balance, real interest rates, and the velocity of base money. The section shows that, despite marked volatility in the different components of the primary balance of the public sector, the balance has been significantly negative throughout the 1990s with the exception of 1995—after a major currency crisis forced adjustment—and 1998, as part of the most recent disinflation attempt. It also shows that the stock of net public debt has doubled during this period in percent of GNP and would have tripled in the absence of inflation. Base money, despite some erosion, is still at 4 percent of GNP by the end of the period. Section B analyzes the relationship between inflation and budget deficits. It develops a formula for the level of the primary balance needed to stabilize the debt-to-GNP ratio in a low inflation environment. Given certain long-term parameters for output growth and the velocity of base money, the chapter shows that a primary surplus of about 2¼ of GNP is needed to stabilize the debt to GNP ratio at 60 percent for a level of the real interest rate equal to 10 percent.

A. Primary Balance, Net Debt, Real Interest Rates, and Real Balances

5. The public sector in Turkey comprises the central government, a set of funds, municipalities and local governments, three social security institutions, 49 state economic enterprises (SEEs), and the central bank.2 Focusing on the balance of the central government, at times, underestimates the extent of the deficit of the entire public sector, as was the case in some years, when the deficit of the rest of the components of the public sector was manifold that of the central government. Moreover, the practice has often been to shift certain expenditures between the accounts of the different segments of the public sector, so that an improvement in one account simply becomes reflected in a worsening in another account. Typically this has to do with budgetary transfers from the central budget to the other segments of the public sector, which when squeezed, lead to an improvement in the primary balance of the central budget but worsen that of funds and SEEs.

6. Table 1 below shows that the primary deficit of the public sector during the first four years of the 1990s leading to the 1994 currency crisis exceeded 5 percent of GNP on average, of which 3½ percent of GNP deficit was at the level of the SEEs.3 After the short-lived adjustment effort in 1995, where a primary surplus of 2¾ percent of GNP was achieved, the primary balance moved into a deficit again in 1996, exceeding 2 percent of GNP in 1997. In early 1998, the government launched a three-year adjustment program with the aim of bringing inflation down to single digits by end-2000 from over 90 percent at end-1997. An improvement in the primary balance of the public sector of 2½ percent was achieved in support of the disinflation strategy, taking the balance to a surplus of ½ percent of GNP. In the run-up to the April 1999 elections the stance of fiscal policy was relaxed significantly. This together with the fiscal cost of the devastating earthquake that hit Turkey in midsummer is likely to lead the primary balance of the public sector to swing back into a deficit of about 2¾ percent of GNP.

Table 1:Turkey: Primary Balance of the Public sector(In percent of GNP)
Primary balance of the public sector-3.6-6.2-7.0-5.6-0.22.7-1.2-2.10.5-2.8
Central government1.3-0.6-1.7-
Extrabudgetary funds-0.6-1.0-0.8-1.5-1.5-0.6-
Local authorities-0.1-0.3-0.1-0.6-0.10.0-0.1-0.1-0.4-0.8
Stale economic enterprises4.2-4.0-4.0-1.9-0.31.3-0.1-0.4-1.1-1.2
Social sec inst. & revolving funds0.0-0.3-0.3-0.6-0.6-0.6-0.20.0-0.4-0.1
Primary component of unpaid duty losses0.00.00.0-0.3-1.1-0.8-1.9-1.4-1.2-1.2

7. Table 1 adjusts the primary balance of each component of the public sector (as defined by the authorities) by the amount of interest receipts that are recorded in their respective accounts as revenues. This adjustment is needed in order to isolate the “truly” primary position, namely, the balance of those revenues and expenditures that are not affected by movements in interest rates. Similarly, the primary balance of the central government is also adjusted for profit transfers from the central bank, as these are also dependent to a large extent on interest rate movements.

8. Table 1 shows that the central government budget was not the major culprit in the extremely weak primary balance during 1990–93, and that it has remained in surplus since 1994 (except in 1997). Over the 1990s the combined primary deficit of the local governments and municipalities4 has been on average equal to 0.3 percent of GNP, as many sources of their financing have been subject to central government control. In 1999 the primary deficit of the local authorities is likely to witness a deterioration (-0.8 percent of GNP) resulting from the large wage increase granted to public sector workers.5 The primary balance of the social security institutions has also been small, as most of their widening deficit was financed through a budgetary transfer from the central budget.

9. A more important source of deficits was the extrabudgetary funds (EBFs), the SEEs, and the unpaid duty losses of state banks.

Extrabudgetary funds

10. In the mid-1980s a proliferation of extrabudgetary funds over which the central government had very little control contributed to a weakening of fiscal discipline.6 In 1993, a major step to restore fiscal discipline in this area was taken when 62 funds, accounting for an estimated 85 percent of total funds revenues, were integrated into the budget. Since then, all of their earmarked tax revenues are shown as part of nontax revenues of the central budget and a budget allocation is decided every year to cover their current commitments. The funds retained, however, the ability to borrow to finance capital spending. Thirteen funds were left and, still are, outside the budget; of these, only the defense fund and the social solidarity fund are large, while the rest are small funds that are not contributing to the overall deficit of the public sector.7

11. In the authorities’ and IMF presentation of the public sector balances, 13 funds are covered under the rubric “extrabudgetary funds.” Eight are budgetary funds—part of the 62 funds that have been integrated into the budget since 1993.8 Two are nonbudgetary funds—the defense fund and the social solidarity fund. Two are newly created funds: the health and education tax fund and the petroleum consumption tax fund. The last “fund” is in fact the budget of the privatization authority, an administration that was set up in the mid-1980s to manage the privatization program. These funds account for about 80 percent of all funds revenues and include all those whose activities that could generate a deficit in addition to that of the central government.

12. As can be seen in Table 1, the primary deficit of the extrabudgetary funds has improved since 1993–94, and has been roughly in balance over the last three years.9

State economic enterprises (SEEs)

13. Despite the ambitious reform program of divestiture, demonopolization, price liberalization, and management and personnel reorganization, which was started in 1980 and deepened in 1984, the accounts of the SEEs have been a major contributor to the overall deficit of the public sector throughout the 1980s and until the early 1990s. At present there are 49 nonfinancial SEEs, employing about 500,000 people. Four of these enterprises are acting as the main vehicle for the government in implementing major agricultural support policies (see the chapter on Agricultural Support Policies), and have recently become one of the main sources of deficit in the SEEs sector.

14. During 1990–93, the average primary balance before budgetary transfers of SEEs remained close to 5 percent of GNP (3.5 percent after budgetary transfers).10 However after the 1994 crisis, SEEs’ overall deficits were reduced sharply, under the combined effect of a major decline in real wages and a sharp curtailment of their investment plan. Some of these trends have started to reverse somewhat during the last three years, and the primary balance of the SEEs, after having turned into a small surplus in 1995, has been again in deficit since 1996. This deficit is projected to be some 1¼ percent of GNP, owing mainly to renewed wage pressures,11 resumption of investment spending, and the carrying of costly stocks of surplus agricultural production.12

Primary Balance of State Economic Enterprises

(In percent of GNP)

Unpaid duty losses of the state banks

15. In addition to the fiscal and quasi-fiscal activities run outside the budget by the budgetary and nonbudgetary funds, and by the SEEs, other quasi-fiscal activities have been run on behalf of the government by the two largest state banks, Ziräat Bank and Halk Bank.13 The main quasi-fiscal activity in which both banks have been engaged is in the provision of subsidized credits to farmers, in the case of Ziräat, and small and medium-scale businesses, in the case of Halk. In addition, Ziräat Bank has acted as an agency to the treasury in collecting taxes and paying salaries to civil servants and public sector workers.

16. Up until around 1994–95, the cost of these quasi-fiscal activities has been borne by these banks and covered by their own profits. Up until that time both banks had a large share of the market both for deposits and commercial lending,14 and the spread between their cost of borrowing and the subsidized rate was relatively limited. Under the impetus of financial liberalization, both banks started to lose market share. Moreover, a large wedge started to appear between their rising cost of funds and the subsidized rate at which they were lending, which was not adjusted commensurately. Their profitability quickly eroded and large deficits started to build.

Ziräat Bank Cost of Funds and Subsidized Interest Rates

(Annual Compounded)

Banks Stock of Unpaid Duty Losses

(In billions of U.S. dollars)

Flow of Unpaid Duty Losses

(In percent of GNP)

17. These deficits were filled by an accumulation of claims on the government (the so-called unpaid duty losses), the yield of which was de facto set so as to cover any loss accumulated during each year.15 This arrangement had two serious drawbacks: first, it did not allow a regular flow of compensation payments to these banks, thus making them increasingly cash strapped, which in turn pushed them to put pressure on interest rates on deposits; second, it resulted in the absence of a clear budget constraint on these banks, as any shortfall of revenues over expenses was covered by the treasury. All these have led to a rapid buildup in these unpaid duty losses, the flow of which is estimated to almost double to over 8 percent of GNP in 1999, bringing the stock close to 12.5 percent of GNP, or close to half of the stock of government’s cash debt.

18. Given the lack of transparency in the way these banks have performed quasi-fiscal operations on behalf of the government, it is hard to analyze and quantify the nature of the losses that have been booked as unpaid duty losses. Analytically, the flow of unpaid duty losses could be decomposed into three components: (i) the credit subsidy on the outstanding stock of subsidized credits, (ii) inefficiency losses, due to overbranching, overstaffing, and reported poor treasury management, and (iii) interest payments on the outstanding stock of unpaid duty losses. The first two components could be looked at as additional primary expenditures of the public sector. The third should be added to interest payments on the stock of public debt. It is very difficult to come up with an estimate of inefficiency losses; accordingly, in this chapter we derive an estimate of credit subsidy and treat the residual flow of unpaid duty losses as interest payments. By doing so we are underestimating the additional primary expenditures of the state banks.

19. There are arguably a number of ways to measure the rate of credit subsidy. For each of these banks one could define the rate of credit subsidy as the difference between their purely commercial lending rates and the subsidized credit rate. However, given the thinness of their purely commercial lending activities, and the lack of a clear benchmark for commercial lending rates in the banking sector, it is preferable to measure the rate of credit subsidy with reference to a different benchmark. The average treasury bill rate is used here. Thus, we measure the amount of credit subsidy as the rate of credit subsidy times the volume of outstanding subsidized credits where the rate of subsidy is the difference between the average treasury bill rate (plus a 10 percent margin to capture credit risk and intermediation costs) and the subsidized interest rate. According to this measure, credit subsidies have been in a range of 1¼ to 1½ percent of GNP in recent years.

Net debt of the public sector and real interest rates

20. As reflected in Table 2 below, the net debt of the public sector in Turkey has increased from less than 30 percent in 1990 to close to 45 percent in 1998 and is projected to reach 58 percent by the end of 1999. This upward trend raises the prospective burden of interest payments, all the more so because it is accompanied by a fall of the share of foreign debt and a marked increase of the share of cash debt, a debt on which the interest rates paid are much higher than on foreign debt and noncash debt.

Table 2.Consolidated Net Debt of the Public Sector(In percent of GNP)
A. Central government plus central bank23.628.530.130.737.834.539.537.538.948.9
Central bank net assets0.0-
Central government31.835.537.438.845.342.848.048.448.560.5
Central government debt held by the CBT8.
B. Rest of the public sector5.
Foreign debt5.
Net domestic dept0.20.7-0.4-1.30.1-0.20.5-0.5-0.52.4
C. Net debt of the public sector28.835.235.735.144.741.344.541.944.588.0
Net foreign debt23.126.525.225.730.729.126.022.520.319.4
Net domestic debt5.68.610.69.414.012.220.420.424.138.6
Of which: cash debt of the central government6.

21. Average ex post real interest rates have been high and volatile with a clear upward trend during the 1990s reflecting the increased share of cash debt. This ex post average real interest rate is estimated to soar to 25 percent in 1999, owing mainly to the government’s resolve to rein in inflation despite nominal interest rates having climbed to close to 140 percent following the Russia crisis in midsummer 1998 and having stayed above 100 percent for most of the year in 1999.16

Net Debt of the Public Sector and Average Ex Post Real Interest Rate

22. Starting from a level of 32 percent at end 1990, the debt-to-GNP ratio would have reached almost 100 percent if inflation had been zero, and if growth, the primary surplus and the real interest rate had been equal to their average values during this period (4.3 percent, 2.6 percent of GNP, and 12.8 percent respectively).17 Of this increase, about 33 percentage points is explained by the primary deficit, and about 35 percentage points to real interest rates in excess of the real GNP growth rate.(See figure below).

Simulation of Debt Dynamics in the Absence of Inflation

23. Table 2 also shows that most of the increase in total debt came from an increase in the debt of the central government (including the duty losses of state banks), whereas that of the rest of the public sector has remained more or less stable, despite the large primary deficits it has run. This is mainly due to the fact that quite often, EBFs, SEEs, and local authorities are unable to roll over their debt and the central budget makes the repayments fallen due on their behalf as most of their debt is contracted with a government guarantee.

Real money balances

24. Table 3 and the figure below show that as inflation ratcheted up during the 1990s, the demand for real base money dropped in percent of GNP. On average, the growth rate of real balances was below 2 percent during 1987–98, while the economy registered a 5 percent growth rate on average during the same period. As a result base money as a share of GNP (the base of the inflation tax) continued to decline falling from over 10 percent in the early 1970s to 6½ percent in 1986 and further to 4½ in 1998. Seignorage during that period averaged 2¾ percent of GNP. Compared to other countries which had gone through episodes of high inflation, Turkey is quite unique in the slow pace at which the demand for real base money has been eroded by inflation, a phenomenon which had prevented the chronic inflation from degenerating into hyperinflation as it did in other countries.

Table 3.Turkey: Real Base Money, Inflation and GNP Growth
Base money 1/Annual percent changeSeignorage

(in percent of GNP) 3/
Real (in trillion 1995 TL) 2/In percent of GNPGNP deflatorReal GNPReal base money

87 to 98

Base money is defined as currency issued plus banks’ required reserves at the central bank.

Base money deflated by the GNP deflator.

Defined as the nominal variation of base money during the year divided by nominal GNP.

Base money is defined as currency issued plus banks’ required reserves at the central bank.

Base money deflated by the GNP deflator.

Defined as the nominal variation of base money during the year divided by nominal GNP.

Index of Real Base Money


Base Money (In percent of GNP) and Inflation 1/

B. The Link Between Inflation and Budget Deficits

25. The simple explanation given in the literature about the relationship between inflation and government deficits views money creation (and the ensuing inflation) as a way to finance the gap between government expenditures and tax revenues, by adding the inflation tax to regular taxes. This “implicit” tax accrues to the government in the form of higher central bank profits, which are transferred directly in the form of dividends or indirectly in the form of interest-free loans.

26. A simple framework in which the relationship between debt and inflation could be formalized is one that looks at the consolidated accounts of the government/central bank. In this framework, the substitutability between debt financing and money financing can be seen, even if, as in Turkey, the government finances itself only through debt creation. In this framework, the increase in the nominal stock of government debt is identically equal to the budget deficit, independently of the level of money creation. However, by printing more money the central bank accumulates larger assets, and hence the increase in the nominal net debt of the consolidated government/central bank is lower than otherwise for the same nominal level of the budget deficit. In this framework it remains true that inflation helps stabilize the government debt-to-GNP ratio itself, and the only channel through which this could be done, with primary deficits and real interest rate higher than the growth rate, is through one form or another of transfers of seignorage revenues to the government. In Turkey, transfers from the central bank to the government that have been reflected in the budget were insignificant. The main form in which revenues from seignorage have been transferred to the government is in the form of cancellation of government debt vis-à-vis the central bank—which are not reflected in the budget, interest-free short-term advances, and lower than market interest rates on noncash government securities held by the central bank.

27. The framework is as follows:

LetBGPRbe the stock of government bonds held by the private sector,
BGFXbe the stock of government bonds held by the foreign sector,
BGCBbe the stock of government bonds held by the central bank,
IGPRbe interest payments on government bonds held by the private sector,
IGFXbe interest payments on government bonds held by the foreign sector inclusive of valuation losses,
IGCBbe interest payments on debt vis-à-vis by the central bank,
TRCBGbe transfers of profits from the central bank to the government, and
Pbe the primary balance of the government excluding transfers from the central bank.

Then the government budget constraint is written as

In return for issuing base money, M, the central bank can acquire claims on either the private sector, CCBPR, the foreign sector, CCBFX, or the government, CCBG or BGCB.

The changes in the central bank balance sheet can thus be written as

where NW is the central bank net worth.

The change in the central bank net worth is its profit. Thus, if we let

IFXCB be interest receipts on the central bank’s net foreign assets CCBFX (inclusive of valuation gains),

IPRCB be interest receipts on the central bank’s claim on the private sector CCBPR, and OP be the operating costs of the central bank, then:

Substituting this expression of ΔNW into (2) and subtracting (2) from (1) we get

If we define D as BGPR+BGFXCCBFXCCBFXCCBPR as the net debt of the consolidated government/central bank, and IP as net interest payments on the net debt (IP=IGPR+IGFXIFXCBIPRCB),

we get the familiar budget constraint of the consolidated government/central bank:18

which simply states that to finance the primary deficit plus net interest payments to the private sector, the consolidated government/central bank needs to have recourse to either bond financing or money financing (seignorage). Both bonds and money are liabilities vis-à-vis the private sector/foreign sector. The former is interest bearing, the latter is noninterest bearing. By altering the mix between bonds and money toward more money, the fiscal/monetary authorities are trading of flower debt burden against higher inflation.

28. The relationship between the primary balance, debt dynamics, and inflation can be derived in continuous time as follows:

Let k=DY be the debt to GNP ratio.


Taking the relationship specified in (4) in continuous time, we have dD = –Pc + iD – dM with Pc the primary balance plus the operating costs of the central bank and t the nominal interest rate.

Substituting this expression into (5) we get

where p=pcY in the consolidated primary balance in percent of GNP; and g is the growth rate of nominal GNP.

Assuming a unit elasticity demand for money such as

Md = Yf(π),

where π is the inflation rate, and differentiating with respect to time and dividing by Y yields:

In steady state dπ = 0 yielding:

where y is growth rate of real GNP. This familiar relationship shows that seignorage dMY is equal to the inflation tax πMY plus a revenue, which is due to the increase in the public’s money holding stemming from growth, yMY.

In steady state MY,, the inverse of base money velocity, is constant. Call it 1ν.

Substituting (7) into (6) yields:

where r is real interest rate.

Equation (8) shows that the debt dynamics are of particular concern when the real interest rate is higher than the growth rate of the economy. In this case the term (r-y)k is positive and requires the sum of the two other terms to be negative to offset it to keep the debt to GDP ratio from rising.

Equation (8) allows us to derive the level of inflation that stabilizes the debt to GDP ratio for any given level of the primary balance. Setting dk = 0 in equation (8) gives:

29. The larger the primary balance, the lower is the inflation needed to stabilize the debt-to-GNP ratio. The larger the gap between the real interest rate and real growth, the higher is the required inflation for the same level of the primary balance. The higher the stock of debt, the higher is the required inflation.19

30. The case of hyperinflation could be seen as a limiting case of this equation. With real interest rate higher than the growth rate and the primary balance in deficit the debt/inflation dynamics could degenerate into hyperinflation. As inflation accelerates, the velocity of base money goes to infinity and thus the required inflation to stabilize the debt-to-GNP ratio becomes infinitely large. In equation (9) π goes to infinity when v goes to infinity.

31. The chart below depicts the relationship between inflation and the value of the primary balance for a level of the real interest rate equal to 10, a debt-to-GNP ratio equal to 60 percent, a growth rate equal to 5 percent, and velocity of base money equal to 16, that is, a ratio of base money to GNP equal to 6.25. A primary deficit of 2.6 percent of GNP would require an inflation of 125 percent to stabilize the debt to GNP ratio. To achieve zero inflation a primary surplus of 2.5 percent of GNP is needed.

Required Inflation as a Function of the Primary Surplus

(y=0.05, v=16, d=.6, and r=.1)

32. In the steady state depicted by equation (9), not only is the net debt-to-GNP ratio constant but also the government debt-to-GNP ratio, Recalling the definition of net debt and the identity representing the balance sheet of the central bank, it is easy to show that stabilizing the net debt-to GNP ratio is consistent with stabilizing the government debt-to GNP ratio as well, as M/Y is constant. With positive real interest rate and negative primary balances the only way in which the government debt-to-GNP ratio could be stabilized is through positive transfers from the central bank to the government.

33. Table 4 shows some sensitivity analysis about the required level of primary surplus that is consistent with stabilizing the debt-to-GNP ratio with an inflation rate of 5 percent. The central scenario is a scenario with a growth rate of 5 percent, a real interest rate of 10 percent, a base money to GNP ratio of 6¼ percent. In this scenario a primary surplus of 2.2 percent of GNP is needed to stabilize the debt-to-GNP ratio at 60 percent. An additional I percent of primary surplus would be required for 2 percentage points of higher real interest rates. An additional 0.2 percent of GNP primary surplus would be needed for each 2 percentage points of lower base money to GNP ratio. An additional 0.6 percent of GNP primary surplus for each 1 percentage point of lower inflation. A 0.5 percent lower primary surplus would be required for a stock of debt 10 percentage points lower.

Table 4.Turkey: The Primary Surplus Needed to Stabilize the Debt to GNP Ratio(With 5 Percent Inflation)
Real growth














Table 5.Turkey: Primary and Overall Deficits of the Consolidated Public Sector, 1990–99(In trillions of Turkish lira)
Primary balance of the public sector-14-39-77-113-6213-187-622255-2,329
Central government (adjusted)5-4-19-12133268184-651,889717
Interest receipts 1/29104071153178
Profits transfers from the CBT0.
Rest of the public sector-19-35-58-95-9812-82-160-1,020-2,043
Extrabudgetary funds-2-6-9-31-58-45-2722-18-387
Interest receipts3572120188
Local authorities-1-2-2-12-53-15-43-209-628
Interest receipts268173269145
SEEs (adjusted for interest receipts)-17-26-44-39-1099-14-127-587-963
Adjusted (before budgetary transfers)-19-39-54-70-3544-75-293-844-1,512
memo: Budgetary transfers2131031255561166258549
Interest receipts458102652118137369658
TMO Merest payments33244423
SSIs and revolving funds0-2-4-13-25-45-26-12-207-65
Interest receipts11112210153051
Primary component of unpaid duty losses000-6-41-67-289-397-614-1,003
Memorandum item:
GNP (in trillion Turkish Lira)3976341,1041,9973,8887,85214,97829,39353,01383,124
Source: Data provided by the Turkish authorities.
Source: Data provided by the Turkish authorities.
Table 6.Turkey. Consolidated Net Debt of the Public Sector, 1990–99(In trillions of Turkish lira)
A. Central government plus central bank941613336141,4682,7135,91211,00920,64440,633
Central bank net assets0-11411-36994542,2624,2358,558
[in million of U.S. dollars]-449-871,6011,164-5752,1965,8468,7489,29614,198
Free reserves in FX1247252178223422761
[in million of US. dollars]5104744944588413577261,0881,3481,425
Claims on banks5491711-4-447281,8382,000
Free reserves in Lira234667533418159
Central government1272254127741,7623,3587,19514,22225,70850,305
[in million of U.S. dollars]23,65925,13425,79828,33630,41631,09532,37531,49932,26332,518
Cash debt2442831373387331,9244,6429,51219,809
Noncash debt010451182064108541,1042,1012,722
Unsecuritized debt vis-à-vis the CBT33466610225521837133800
Unpaid duty losses00013751646331,5233,98110,394
Deposits at Ziraat361176334-25-25
Central government debt held by the CBT3346661393305478289818281,014
Unsecuritized debt vis-à-vis the CBT33466610225821837133800
B. Rest of the public sector214262872685281,0471,5872,9207,586
Foreign debt2038671142645449701,7273,1795,607
Total (in millions of U.S. dollars)6,7567,4477,7797,8779,0279,1429,0218,43410,16710,505
Local authorities1,0321,1591,6951,6021,5492,4142,7392,6892,7552,745
Net domestic debt14-5-275-1677-139-2591,979
Revenue sharing certificates201301559
Other (bank net debt) 1/-11-14-21-39-38-50-63-429890
Local authorities (bank net debt) 1/-1-2-3-5-7-14-53-108-172
Commercial bonks12201317405141205415
Bank deposits 1/388122253145207428
C. Net debt or the public sector1142233947011,7313,2406,95912,59623,56548,218
Memorandum items:
Adjusted eop TL/US$2,9305,0308,55614,45829,20559,501107,505204,750312,720
Source: Data provided by the Turkish authorities.
Source: Data provided by the Turkish authorities.
Table 7:Turkey: Average Ex Post Real Net Interest Rate on Public Domestic, 1990–99(In trillions of Turkish liras, unless otherwise specified)
Net interest payments16.232.659.7148.2391.7711.81,786.03,245.48,687.417,126.4
Monetary correction0.017.327.859.2187.8383.9902.42,991.05,844.08,069.4
Real interest payments16.215.431.989.0204.0328.0883.5253.42,843.49,057.0
Net debt of the public sector1142233947011,7373,2406,95912,59623,56548,218
Real interest rate14.29.110.316.316.713.217.32.615.725.2
Source: Data provided by the Turkish authorities.
Source: Data provided by the Turkish authorities.
Table 8.Turkey: Simulation of Debt Dynamics, 1990–99
Primary deficit-2.7-2.8-2.9-3.0-3.2-3.3-3.4-3.6-3.7
in percent of GNP-2.6-2.6-2.6-2.6-2.6-2.6-2.6-2.6-2.6
Interest payments4.
interest rate12.812.812.812.812.812.812.812.812.8
in percent of GNP31.837.041.648.655.162.269.878.187.096.7
An alternative scenario with lower real interest rate
Interest payments1.
interest rate4.
in percent of GNP31.834.437.039.542.144.647.249.752.354.8
An alternative scenario with zero primary deficit
Interest payments4.
interest rate12.812.812.812.812.812.812.812.812.8
in percent of GNP31.834.537.340.343.647.
Memorandum items:
GNP growth rate4.
Source: Fund staff calculations.
Source: Fund staff calculations.

Prepared by Rakia Moalla-Fetini.

Financial public institutions (four money deposit banks and three investment and development banks) are not formally part of the consolidated public sector in the IMF definition. However, the losses run by the two main public banks—stemming from quasi-fiscal operations that they are performing on behalf of the government—are included in the consolidated public sector account, as discussed in this chapter.

Background data for this chapter are provided in Tables 58.

There are 80 provinces, 15 large metropolitan municipalities, and 2,827 smaller municipalities.

In early 1999, new two-year contracts were renegotiated with public sector workers, providing for an up-front increase of 39 percent plus three consecutive six monthly adjustments equal for the past six month CPI inflation times 1.05. Due to these contracts, public sector worker wages are expected to increase by more than 20 percent in real terms in 1999.

As part of the austerity measures taken in 1988, a clawback provision on the funds’ revenues was introduced, whereby the funds had to transfer a share of their earmarked tax revenues to the central budget. This measure had constrained somewhat the profligacy of funds—as they did not have the ability to borrow other than to finance capital spending—but it did not achieve a major improvement, in the absence of any centralized control on their operations.

These funds include the insurance fund for time deposits, the capital market board fund, the insurance inspection board fund, and the highway traffic insurance fund.

These are the mass housing fund, the public participation fund, the price stabilization fund, the revenue administration improvement fund, the support and price stabilization fund, the resource utilization and support fund, the oil exploration fund, and the fuel price stabilization fund.

Two adjustments are made to the authorities’ measurement of the primary balance of the extrabudgetary funds. Net lending by the mass housing fund is considered as an expenditure, while in the authorities’ presentation it is treated as a financing item. Repayments of loans extended by the price stabilization and support fund are treated as a negative expenditure item, while in the authorities’ definition they are treated as a financing item.

In addition to transfers from the budget in the form of equity, compensation for duty losses, and aid, the SEEs receive other subsidies and transfers such as those from the support and price stabilization fund to fertilizer producers and transfers from the development and support fund to the electricity company. Other mechanisms of financing SEE deficits were tax arrears and unpaid dividends, and unpaid premia to the social security institutions.

The same wage contracts as those covering public sector workers in local administrations cover workers in state economic enterprises.

In Table 67, the primary balance of the SEEs, as shown in the authorities’ presentation, is adjusted downward by the amount of interest receipts and upward in the last three years by interest charges paid by soil products office (TMO) and which have not been classified as such in the consolidated accounts of the SEEs. These interest charges have become quite significant given the large stocks of grain that TMO has accumulated in the last few years.

Chronic liquidity pressures have been addressed from time to time through capital injections and minor payments of the duty losses by the treasury.

Ziräat Bank alone had 20 percent of the commercial lending market as opposed to 7 percent now.

Formally, Halk Bank’s additional flow of unpaid duty losses is supposed to be calculated as interest income on the stock of unpaid duty losses at “commercial rates,” while that for Ziräat Bank was supposed to be calculated based on Ziräat “cost of funds.”

The dip in real ex post interest rate in 1997 is in part due to a lengthening of maturity.

The exceptionally high real interest rate of 25 percent in 1999 is excluded from the calculation of the average real interest rate.

In what follows, domestic and foreign debt is aggregated and k refers to the ratio of domestic and foreign debt to GNP. This abstracts from the effects of a real appreciation (which lower the foreign debt to GNP ratio) since only a modest real appreciation can be sustained over time. See Chapter V.

To apply this formula on actual data requires converting discrete time growth rates-in this case annual growth rate- to continuous time growth. The relationship between both is as follow: In (1 + gd) = gc, where gd and gc are respectively discrete and continuous time growth rates. For example, an annual growth rate of 10 percent translates into a continuous growth rate of 9.5 percent.

Other Resources Citing This Publication