STELLENBOSCH UNIVERSITY

DEPARTMENT OF ECONOMICS

ECONOMICS 348

FISCAL POLICY IN SOUTH AFRICA

Krige Siebrits

(Department of Economics, Stellenbosch University)

Prescribed material prepared for the Economics 348 module

1
 CHAPTER 3
WHEN IS FISCAL POLICY SUSTAINABLE?

3.1 INTRODUCTION AND STUDY OBJECTIVES

Chapter 2 showed that several sound justifications exist for running budget deficits, but
added that the frequency, size and financing of deficits all influence their economic
effects. It then proceeded to discuss four ways of financing governments' budget deficits.
Two of them, namely seigniorage and running down foreign reserves, were shown to be
highly risky and incapable of financing large deficits. Their riskiness stems from ten-
dencies to create macroeconomic problems: seigniorage can cause hyperinflation, while
depletion of foreign reserves can cause balance of payments crises. Chapter 2 also
pointed out that domestic and foreign borrowing are more realistic and prudent ways to
finance budget deficits. But is also warned that excessive borrowing in domestic or
international financial markets can give rise to crushing debt crises and other economic
problems.
This warning raises an obvious question: when is government borrowing "exces-
sive" and likely to cause a public debt crisis? The present chapter discusses this ques-
tion. Because the question is usually asked with prevention of such crises in mind, this
chapter follows most economists by framing it in the following way: when is fiscal policy
sustainable? A preliminary answer was provided in Section 1.3, which stated that a fiscal
position is deemed sustainable when the level of public debt can be maintained, that is,
when the country can service and repay it. The present chapter derives practical guide-
lines from this statement. It will become clear that economists understand the dynamics
of public debt very well but lack clear criteria for assessing the sustainability of fiscal
positions. In the nutshell, this simply reflects that sustainability assessment is very com-
plex, mainly because it involves predicting the future. Nonetheless, this chapter explains
some concepts and methods that are helpful in fiscal sustainability assessment. These
concepts and methods also feature in Chapter 4, which discusses the sustainability of
fiscal policy in South Africa.

After you have studied the contents of this chapter, you should be able to:

• Briefly explain why and how economists assess the sustainability of fiscal
policy in the past.
• Define the terms "fiscal solvency" and "fiscal sustainability".
• Explain the dynamics of the public debt.
• Explain the role of the primary balance in fiscal sustainability assessment.
• Discuss the nature, role and limitations of sovereign credit ratings.
• Discuss empirical research into the relationship between public debt burdens
and economic growth.

31
3.2 PUBLIC DEBT BURDENS AND DEFAULTS

Actual debt crises seem potentially rich sources of information for specifying fiscal sus-
tainability criteria. At what levels of public debt did countries default in the past? Are
there patterns that could be used to derive a threshold beyond which public debt crises
are inevitable? Figure 6 suggests that this is not the case: in the years in which 15 coun-
tries defaulted in recent times their public debt burdens ranged from 21.3% of GDP to
160.0% (the ratios in the years just before the defaults ranged from 22.5% to 172.1%).
This conclusion is reinforced by the experiences of other countries that have maintained
very high public debt burdens for extended periods of time without having debt crises.
Japan, where the public debt-to-GDP ratio has exceeded 150% of GDP in every year since
2001 and reached 247.9% in 2015, is a well-known example.

Figure 6
Public debt burdens before and at the time of sovereign defaults (1999-2016)
Venezuela (2005)
Year before default
Uruguay (2003)
Ukraine (2015) Year of default
Suriname (2000)
Russia (1999)
Paraguay (2003)
Pakistan (1999)
Jamaica (2013)
Indonesia (1999)
Greece (2012)
Ecuador (2008)
Dominican Republic (2005)
Cyprus (2013)
Cameroon (2004)
Argentina (2014)

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

% of GDP
Sources : S&P Sovereign Default Database; International Monetary Fund Historical Public Debt Database.

One reason for the absence of a clear pattern is that country's public debt repayment
and servicing records reflect financial considerations (the ability to pay) as well as poli-
tical ones (the willingness to pay). Governments sometimes default deliberately despite
being able to meet their debt obligations. But the debt burdens of governments that
defaulted solely for financial reasons also varied markedly. The manageability of a given
public debt burden is determined by various factors such as the growth performance of
the economy, levels of interest rates, and aspects of its composition (for example, the
ratios between short-term and long-term debt and between domestic and foreign debt).
This chapter will return to some of these factors. The important point to note is that we
cannot say that a country is heading for a crisis simply because its public debt-to-GDP
ratio has reached a certain level such as 60% or 100%. Meaningful assessment of debt
burdens also requires information about other economic aggregates and judicious use of
sustainability indicators discussed elsewhere in this chapter.

32
3.3 THE SUSTAINABILITY OF FISCAL POLICY IN THE PAST

Many researchers have applied econometric and other methods to establish whether
countries' fiscal positions were sustainable in the past. The question may be asked
whether such research is useful: perhaps all we need to know about the sustainability of
fiscal policy in the past is whether countries defaulted and experienced debt crises or
not. In fact, researchers and policymakers can learn much from proper analysis of fiscal
sustainability in the past. For one thing, such research makes it possible to identify and
explore the requirements for maintaining fiscal sustainability over time. Moreover, time-
ly interventions by policymakers often correct unsustainable fiscal positions before debt
crises can occur. Valuable knowledge can be obtained from identifying and studying
such policy interventions. Chapter 4 uses the findings of ex post sustainability analyses
of fiscal policy in South Africa for both purposes. In preparation for those parts of Chap-
ter 4, the remainder of this section provides introductions to three methods of ex post
fiscal sustainability analysis.
The first two methods involve time-series econometrics. Cointegration analysis
can be used to assess fiscal sustainability by establishing whether certain fiscal aggre-
gates were cointegrated, that is, whether a long-term (or equilibrium) relationship exis-
ted between them. When government revenue and government expenditure were coin-
tegrated, it means that decisions about the two aggregates were not taken in isolation
and that one did not drift too far apart from the other. Hence, statistical tests that show
cointegration between the two aggregates in periods in the past imply that fiscal policy
was sustainable. Another time-series method that is used to assess sustainability ex post
involves econometric estimation of fiscal reaction functions. A fiscal reaction function
specifies the reaction of the primary balance (recall that Section 1.2.3 defined the
primary balance as the difference between the government's total revenue and non-
interest expenditure) to changes in the public debt-to-GDP ratio. When the public debt-
to-GDP ratio increases, the primary deficit that causes the increase must improve in
subsequent years (that is, must become smaller or turn into a primary surplus) to slow
or reverse the accumulation of debt. Hence, fiscal policy was sustainable in a period if
the estimated fiscal reaction function indicates that increases in the public debt-to-GDP
during that period were followed by improvements in the primary balance.
The third method of ex post fiscal sustainability analysis is a case study variant of
fiscal reaction function analysis. Whereas the fiscal reaction function method identifies
the average response of the primary balance to changes in the public debt-to-GDP ratio
over a period of time, the case study method studies such responses one by one. Assume,
for example, that the public debt-to-GDP ratio of a country increased 20 times during the
59 years from 1960 to 2019. One of these increases occurred in 2010. A fiscal reaction
function analysis would span the entire period and its findings would tell us how the
primary balance responded on average to the 20 increases in the public debt burden. By
contrast, the case study approach would provide a detailed analysis – usually in the form
of a narrative – of the response to one of these increases (for example, the postulated
one in 2010).

33
3.4 THE FUTURE SUSTAINABILITY OF FISCAL POLICY

Chapter 2 pointed out that the economic, social and often also the political costs of debt
crises are very high. Hence, the future sustainability of fiscal policy is always a matter of
interest to policymakers. It should be clear, however, that the description of sustainable
fiscal policy used thus far (namely, a level of public debt can be maintained) is of no
practical value for this purpose. Policymakers need a more precise definition of fiscal
sustainability and "early warning" indicators that enable them to identify and correct
fiscal imbalances before debt crises occur. These issues are the topics of the remainder
of this section. Section 3.4.1 presents a definition of forward-looking fiscal sustainability
and explains its complexity. Section 3.4.2 uses the debt dynamics framework to explain
the influence of key drivers of the evolution of the public debt burden. Section 3.4.3
focuses on the value of the primary balance as an "early warning indicator" of emerging
public debt problems.

3.4.1 The complexity of forward‐looking fiscal sustainability analysis

The notion of fiscal sustainability is closely related to that of fiscal solvency. Hence, the
first step towards formulating a more precise definition of fiscal sustainability is to
define fiscal solvency. Debrun, Ostry, Willems and Wyplosz (2020: 153) state that a
government is solvent when it can "honour its current and future financial obligations".
The net worth of firms (that is, the difference between total assets and total liabilities) is
normally used to assess their solvency. But relatively few governments publish balance
sheets with such information. Furthermore, net worth is a problematic concept in the
government sector because it cannot account for important "assets" such as the power
to raise more revenue by introducing new taxes. For these and other reasons, the reve-
nues, expenditures and debts of governments are the building-blocks for specifying their
solvency requirement. Calitz, Du Plessis and Siebrits (2014: 58) formulate the solvency
requirement as follows: "A government is solvent if the present value of the flow of all
future revenues exceeds the value of outstanding debt plus the discounted value of
future government expenditure". This means that the government should run sufficient-
ly large primary surpluses in future to enable it to pay interest on its debt and to repay
the principal. It follows that a fiscal position is sustainable if the fiscal authorities can
maintain the same set of policies indefinitely without becoming insolvent.
Compared to the alternative of describing sustainable fiscal policy as "fiscal poli-
cy that can be maintained over time", a definition that explicitly incorporates the solven-
cy requirement is more informative. But it remains of little practical value to policy-
makers because all future revenues and future government expenditures cannot be
quantified. Note that the definition as such is not the problem; the real constraint is the
nature of fiscal solvency itself. As Debrun et al. (2020: 153) put it: "Because solvency
boils down to a mere prediction about future budget balances over an indefinite horizon,
it has no clear operational implication". They conclude their discussion of fiscal sustain-
ability with the following statement:

34
 Assessing public debt sustainability is as critical as it is complicated. It is
critical because unsustainable debts often end up in some costly combina-
tion of default, high inflation, and a broken financial system. It is complica-
ted because sustainability is inextricably linked to solvency… Thus, sus-
tainability is a purely forward-looking concept, and assessing it amounts
to making a prediction about the unknowable future" (Debrun et al., 2020:
185).
Because it is so important to do so, economists remain engaged in what Wyplosz (2011)
labels "mission impossible" by undertaking fiscal sustainability analysis. While there is
no solution to the problem that such analyses are essentially predictions, their value can
be enhanced by realistic assumptions about the future course of government revenue
and expenditure, economic growth rates and interest rates. In addition, organisations
such as the IMF often prepare many scenarios when assessing the sustainability of coun-
tries' fiscal positions. By presenting the results as fan charts containing distributions
around a baseline, users of the analyses can infer the statistical probabilities of predic-
ted public debt outcomes. The specific purpose of some scenarios of this nature is to
mimic the implications for the sustainability of the public debt of adverse shocks. For
example, a scenario could be run to ascertain the sustainability implications of a global
pandemic that causes severe recessions in the country in question and its major trading
partners, a large decrease in tax revenues and unplanned borrowing to finance emer-
gency spending. It would be indicative evidence of a sustainable fiscal position if the sce-
nario shows that such a severe shock would not make the country's debt burden
unmanageable.

3.4.2 Public debt dynamics

As was suggested earlier in this section, the public debt dynamics framework identifies
key drivers of public debt growth and illustrates their effects on the evolution of the
debt burden. Given the problematic nature of forward-looking fiscal sustainability ana-
lyses, it is useful to supplement their findings with information about emerging trends in
these drivers of public debt growth. The debt dynamics equation can be specified in
several ways. This section uses a specification suggested by Blanchard (2017: 459-460):
(Bt / Yt) – (Bt‐1 / Yt‐1) = (r – g) (Bt‐1 / Yt‐1) + (Gt – Tt) / Yt [3.1]
In this expression, B represents the public debt, Y the GDP, r the real interest rate, g the
real economic growth rate, G the non-interest expenditure of the government, and T the
government's revenue. The subscripts "t" and "t‐1" refer to the current fiscal year and
the previous fiscal year, respectively.
Equation 3.1 is much simpler than it may appear to be. The left-hand side of the
expression simply represents the change in the public debt-to-GDP ratio during a given
year, that is, the ratio at the end of year t minus the ratio at the end of year t‐1. The right-
hand side has two terms that jointly determine the change in the public debt-to-GDP
ratio. The first is the interest component of government spending, also expressed as a
ratio of GDP. It is driven by two factors: the real interest rate and the real rate of econo-

35
mic growth. This subsection and the next one will explain the significance of these two
factors. The second term on the right-hand side of the equation represents the primary
balance-to-GDP ratio (that is, the difference between total revenue and non-interest
expenditure, expressed as a percentage of GDP). Hence, the debt dynamics equation
ascribes changes in the public debt-to-GDP ratio to two factors: the government's pri-
mary balance and its interest bill. Trends in three variables drive these trends: the initial
public debt-to-GDP ratio, the real interest rate, and the real rate of economic growth.
The following statements describe the relationships between these factors and increases
in the public debt-to-GDP ratio:
 All other things equal, the higher the initial public-debt-to-GDP ratio, the larger
the increase in the public debt-to-GDP ratio will be.
 All other things equal, the higher the real interest rate, the larger the increase in
the public debt-to-GDP ratio will be.
 All other things equal, the slower the real economic growth rate, the larger the
increase in the public debt-to-GDP ratio will be.
 All other things equal, the larger the primary deficit, the larger the increase in the
public debt-to-GDP ratio will be.
These relationships are useful for deriving "early warning" indicators. For example,
assume that the real interest rate in a country is increasing and likely to remain rela-
tively high in the foreseeable future. In addition, the country's growth prospects are
poor. Since both these developments would put upward pressure on the public debt-to-
GDP ratio, policymakers would be well advised to avoid primary deficits (let alone large
ones) to prevent the emergence of an unsustainable fiscal position.

3.4.3 The critical importance of the primary balance

The debt dynamics framework reveals that the primary balance is a key driver of public
debt growth and, hence, crucial for fiscal sustainability assessment. But the role of the
primary balance in fiscal sustainability assessment can be approached from a somewhat
different angle as well: it is a measure of the government's ability to service its debt from
its ordinary revenue.23 This makes the primary balance an important measure of the
effect of the budget on the government's net indebtedness.
Using the primary balance for this purpose involves comparing it to the govern-
ment's interest bill. Three possibilities exist: a primary surplus larger than the interest
bill, a primary surplus smaller than the interest bill, and a primary deficit. The three
examples in Table 6 can be used to explain the effects of these possibilities on the net
indebtedness of the government in the following way:
 The implication of running a primary surplus larger than the interest bill is that
the government's ordinary revenue can finance all its interest and non-interest
expenditure and redeem some of its debt as well. The second column in Table 6
illustrates this scenario. Here, a primary surplus of R23 billion remains after non-

23 Recall that the ability to service debt refers to the ability to make the required interest payments.

36
 interest spending of R224 billion had been defrayed from revenues totalling
R247 billion. This surplus is sufficient to pay the interest bill of R19 billion and to
redeem R4 billion of public debt. The government's net indebtedness therefore
decreases when it runs a primary surplus larger than the interest bill.
 When the government runs a primary surplus smaller than the interest bill, its
ordinary revenue can finance its non-interest expenditure and a portion of its
interest bill. However, the remaining portion of the interest bill is capitalised and
added to the public debt. In the example in the third column in Table 6, the total
revenue of R247 billion is sufficient to finance all non-interest spending (which
amounts to R232 billion) as well as R15 billion of the interest bill. The remaining
R4 billion of the interest bill is added to the public debt. The government's net
indebtedness increases in this scenario.
 The example in the third column in Table 6 shows that it is an unsound fiscal
practice to run primary deficits. In this scenario, the government's revenue
equals R247 billion and its total expenditure R271 billion. The non-interest por-
tion of total expenditure amounts to R252 billion. The implication is that the
government must incur debt to finance R5 billion of non-interest spending (the
primary deficit) and the full interest bill of R19 billion. Increases in net indebted-
ness of this nature sometimes initiate periods of rapid growth in public debt-to-
GDP ratios.

Table 6
The primary balance and the government's net indebtedness: Hypothetical examples
Aggregate Primary surplus Primary surplus Primary deficit
larger than the smaller than the
interest bill interest bill
(Rand billion) (Rand billion) (Rand billion)
Total revenue 247 247 247
Interest expenditure 19 19 19
Non-interest expenditure 224 232 252
Total expenditure 243 251 271
Primary balance 23 15 -5

These simple examples make it clear that governments should avoid primary deficits. It
follows that the emergence of a primary deficit is an obvious "early warning" indicator:
when that occurs, the fiscal authorities should restore a primary surplus to prevent
more serious debt problems.24 However, a very important result known as the "primary
balance rule of thumb" implies that primary deficits do not always lead to increases in
public debt to GDP ratios. It states that primary deficits can be maintained for some time
without an increase in the government’s debt-to-GDP ratio, provided that the real rate of
economic growth is higher than the real rate of interest. The primary balance rule of

24Section 3.3 showed that the purpose of fiscal reaction function analysis is to establish whether they did
so in the past.

37
thumb follows from the debt dynamics equation. Recall that the right-hand side of that
equation contains two terms that drive changes in public debt-to-GDP ratios. The first of
these is the primary balance-to-GDP ratio. The second describes the evolution of the
interest expenditure-to-GDP ratio, and is formulated as follows:
(r – g) (Bt‐1 / Yt‐1) [3.2]
If the real economic growth rate ("g") is higher than the real interest rate ("r"), the
interest expenditure-to-GDP ratio decreases, all other things equal. The primary balance
rule of thumb implies that a sufficiently large positive differential between the real eco-
nomic growth rate and the real interest rate can induce a decrease in the interest
spending-to-GDP ratio that is big enough to prevent a primary deficit from boosting the
public debt-to-GDP.25 The implication of this result is that economic growth is a power-
ful mechanism for restraining growth in public debt-to-GDP ratios.
It should be emphasised that the primary balance rule of thumb does not imply
that governments can run primary deficits indefinitely. It simply means that they have
some time to eliminate the primary deficit if the real economic growth rate exceeds the
real interest rate. But they must restore a primary surplus to prevent debt problems.

3.5 SOVEREIGN CREDIT RATINGS

Credit rating agencies are private firms that assign values to the credit risks of various
kinds of securities, including government bonds. Investors routinely use these ratings
when assessing the riskiness of investing in government bonds and other securities.
Credit risk ratings are essentially measures of default risk. In this sense, they are closely
related to sustainability assessments.
Although many credit rating agencies exist, three firms dominate the market.
These three firms ― Moody's, Standard and Poor's, and Fitch ― hold an estimated 95%
of the market. Table 7 shows their ratings categories. Note the two basic categories:
investment grade and non-investment grade. Bonds with non-investment grade ratings
are often described as "junk bonds". The lowest "junk-bond" categories are reserved for
securities on which the issuers have defaulted.
The credit rating of the three major agencies matter greatly to fiscal policymakers
and to the citizens of countries. Poor credit ratings constrain fiscal authorities' access to
funding because the rules of some funds prohibit investments in non-investment grade
bonds. Hence, countries usually experience capital outflows when ratings agencies
downgrade their government bonds to non-investment grade status, because at least
some foreigners sell their holdings. Furthermore, the riskier a security is deemed to be,
the higher the interest rate the issuer has to offer to induce investors to buy it. Hence,
poor credit ratings also make the interest rates on loans significantly higher. For govern-
ments, this means a higher interest bill and fewer resources to spend on preferred items
such as education, healthcare, and infrastructure.

25 Put differently, (r – g) (Bt‐1 / Yt‐1) can reduce (Bt / Yt) – (Bt‐1 / Yt‐1) by more than (Gt – Tt) / Yt increases it
if g is sufficiently larger than r.

38
 Table 7
Ratings categories of the three main rating agencies
Fitch Moody's S&P Explanation
AAA Aaa AAA Prime
AA+ Aa1 AA+
AA Aa2 AA High grade
AA- Aa3 AA-
Invest-
A+ A1 A+
ment
A A2 A Upper medium grade
grade
A- A3 A-
BBB+ Baa1 BBB+
BBB Baa2 BBB Lower medium grade
BBB- Baa3 BBB-
BB+ Ba1 BB+
BB Ba2 BB Non-investment grade speculative
BB- Ba3 BB-
B+ B1 B+
B B2 B Highly speculative
Non invest-
B- B3 B-
ment
CCC+ Caa1 CCC+ Substantial risks
grade
CCC Caa2 CCC Extremely speculative
("Junk")
CCC- Caa3 CCC-
In default with little prospect for recovery
CC+ Ca CC
CC C
CC- In default
DDD D D

Because of this, politicians and government officials sometimes react angrily

when rating agencies assign poor ratings to their securities or downgrade them to "junk
status". Governments, however, should not target the messengers when they mismanage
their finances. It should also be kept in mind that the rating agencies' primary respon-
sibility is to investors, not to governments and other issuers of securities. Many invest-
ment funds manage the life savings of ordinary people and rely in part on the rating
agencies to avoid bad decisions that could ruin their reputations and the personal finan-
ces of their clients.
Nonetheless, rating agencies should be held accountable for the accuracy of their
risk assessments. The major agencies are well-resourced, employ highly skilled staff and
have well-established techniques for assessing credit risk. Sovereign risk assessment,
however, is no less complicated than fiscal sustainability assessment. In recent years, the
three major credit rating agencies have been criticised heavily for a series of major blun-
ders. In the best-known cases, such as those of the US firm Enron in 2001 and the US
housing-finance providers Freddie Mac and Fannie Mae during the subprime crisis in
2007 and 2008, they assigned prime investment grade ratings to entities right until they
collapsed. The problem seems to be that investors typically use the ratings of all three,

39
and that these ratings tend to be very similar. Hence, there is not really effective compe-
tition between the agencies, which means that clients cannot hold them accountable for
errors by switching to rivals. But until a solution is found for this problem, no govern-
ment can ignore the effects of credit ratings on their finances and the welfare of their
countries' citizens.26

3.6 PUBLIC DEBT AND ECONOMIC GROWTH

Another vital question linked to the issue of fiscal sustainability is: how do high levels of
debt affect the performance of economies? Section 2.4.2 used the example of Greece to
show that public debt crises usually affect economies very negatively. But is it possible
to identify a point before a debt crisis erupts at which public debt begins to hamper eco-
nomic performance? Although it is possible to do the same for other key economic out-
comes (such as inflation and the incidence of poverty), this section focuses on the rela-
tionship between public debt burdens and rates of economic growth. This relationship
matters for two reasons. The first, and most obvious, is that economic growth is a critical
determinant of general improvements in living standards. The second follows from the
discussion of the debt dynamics frameworks in Section 3.4.2. That discussion showed
that rapid economic growth slows growth in public debt-to-GDP ratios, all other things
equal. This hints at the possibility of a vicious circle: an increasing public debt-to-GDP
ratio may well be self-reinforcing if it prevents one of the main brakes to such increases,
namely rapid economic growth.
Most studies of the link between levels of public debt and rates of economic
growth have reported a negative relationship: high levels of public debt seem to under-
mine economic growth. A paper by Rogoff and Reinhart (2010) is one of the best-known
of these studies. Figure 7, which contains two charts, depicts their main findings for the
period 1946-2009. The one on the left contains results for 20 high-income countries,
while the one on the right reports results of a similar analysis of 24 emerging market
economies. The four columns in each chart show the average real growth rates of coun-
tries in years during which their debt burdens fell into the following categories: smaller
than 30% of GDP, between 30% and 60% of GDP, between 60% and 90% of GDP, and
more than 90% of GDP. On average, high-income countries had the highest GDP growth
rates (4.1% per annum) when their debt burdens were smaller than 30% of GDP. Debt
burdens between 30% and 60% of GDP as well as those between 60% and 90% of GDP
were associated with average annual growth rates of 2.8% in real terms. Strikingly, the
average real growth rate dropped to –0.1% in years in which countries had public debt-
to-GDP ratios above 90% of GDP. Broadly speaking, the pattern for emerging market
economies was similar, although less dramatic. In those countries, average real econo-
mic growth rates changed from 4.3% per annum when public debt burdens were
smaller than 30% to 4.8% per annum when debt burdens ranged between 30% and
60% of GDP, 4.1% per annum when debt burdens ranged between 60% and 90% of

26 White (2010) provides a more extensive discussion of the roles of credit rating agencies.

40
 Figure 7
Public debt and economic growth (1946-2009)
20 Advanced economies 24 Emerging market economies
5 6

4 5

3 4

Real GDP growth

Real GDP growth

2 3

1 2

0 1

-1 0
< 30% 30% - 60% 60% - 90% > 90% < 30% 30% - 60% 60% - 90% > 90%
Public debt to GDP ratio Public debt to GDP ratio
Source : Reinhart and Rogoff (2010: 25).

GDP, and 1.3% per annum when debt burdens exceeded 90% of GDP. The authors con-
cluded that public debt burdens of up to 90% of GDP do not seem to reduce economic
growth sharply. But the growth costs of public debt burdens seem to become much
larger when the public-debt-to-GDP ratio exceeds 90% of GDP.
Rogoff and Reinhart suffered embarrassment when Thomas Herndon (then a PhD
student at the Massachusetts Institute of Technology), discovered an error in one of
their Excel spreadsheets. When he corrected the error, the average annual growth rate
in high-income countries with debt burdens of more than 90% of GDP was 2.2% in real
terms (not -0.1%). But this finding merely weakened their argument. The basic result
that higher public debt-to-GDP burdens are associated with lower real GDP growth rates
was not overturned.
Moreover, other studies have confirmed Rogoff and Reinhart's findings. In fact,
some have found that large loan liabilities may undermine economic growth long before
public debt ratios reach 90% of GDP. One of these papers (Caner, Grennes and Koehler-
Geib, 2010) studied the relationship between public debt burdens and economic growth
rates in 26 high-income and 75 developing countries from 1980 to 2008. It found that a
public debt burden of 77% of GDP was an important threshold in high-income countries.
Every 1 percentage point increase in the debt ratio above 77% of GDP was associated
with a 0.017 percentage point drop in the real economic growth rate per annum. This
number seems tiny, but small differences in growth rates become significant when com–
pounded over time. The corresponding threshold for emerging market economies was a
public debt-to-GDP ratio of only 64% of GDP. Above that, every 1 percentage point
increase in the debt ratio was associated with decrease of 0.02 percentage point drop in
real economic growth per annum. In general, empirical research does not suggest that
government can raise economic growth rates in a lasting manner by means of debt-
financed fiscal stimulus measures.

41
3.7 IN CONCLUSION

This chapter showed that it is vital to maintain fiscal sustainability for two reasons: to
avoid costly debt crises, and to have better economic performance (for example, higher
rates of economic growth). But sustainability analysis is clearly not an exact science.
Many useful tools exist but none of them gives unfailing results ― largely because sus-
tainability assessment is essentially an exercise in prediction. Because of the lack of pre-
cision of all methods for assessing fiscal sustainability, policymakers should err on the
side of caution by avoiding public debt burdens that could give rise to debt crises or
weak economic performance. In addition, they should use the available "early warning"
indicators to prevent serious debt problems, for example, by taking steps to eliminate
primary deficits. The reason for taking such a cautious approach is that the costs of
maintaining prudent debt levels (such as the need for higher tax burdens and having
somewhat less money available for government spending programmes) tend to be
smaller than the economic, social and political costs of public debt crises.

3.8 REFERENCES

Blanchard, O.J. 2017. Macroeconomics (7th edition). Boston: Pearson.

Calitz, E., S.A. du Plessis and F.K. Siebrits. 2014. Fiscal sustainability in South Africa: Can
history repeat itself? Studies in Economics and Econometrics, 38(3): 55-78.
Caner, M., T. Grennes and F. Koehler-Geib. 2010. Finding the tipping point ― When sove-
reign debt turns bad. Policy Research Working Paper No. 5391. Washington, DC:
The World Bank.
Debrun, X., J.D. Ostry, T. Willems and C. Wyplosz. 2020. Debt sustainability. In Sovereign
Debt: A Guide for Economists and Practitioners (edited by S. Ali Abbas, A. Pienkow-
ski and K. Rogoff). Oxford: Oxford University Press: 151-191.
Rogoff, K. and C.M. Reinhart. 2010. Growth in a time of debt. American Economic Review
Papers and Proceedings 100(2): 573-578.
White, L.J. 2010. The credit rating agencies. Journal of Economic Perspectives, 24(2): 211-
226.
Wyplosz, C. 2011. Debt sustainability assessment: Mission impossible. Review of Econo‐
mics and Institutions, 2(3): 1-37.

42