Swarthmore College

Works

Economics Faculty Works Economics

2012

Inflation Targeting In Africa

Stephen A. O'Connell
Swarthmore College, soconne1@swarthmore.edu

Follow this and additional works at: https://works.swarthmore.edu/fac-economics

Part of the Economics Commons

Let us know how access to these works benefits you

Recommended Citation
Stephen A. O'Connell. (2012). "Inflation Targeting In Africa". Oxford Companion To The Economics Of
Africa. 359-364.
https://works.swarthmore.edu/fac-economics/256

This work is brought to you for free by Swarthmore College Libraries' Works. It has been accepted for inclusion in
Economics Faculty Works by an authorized administrator of Works. For more information, please contact
myworks@swarthmore.edu.
 Inflation Targeting in Africa
Stephen A. O'Connell

1. Introduction
In an inflation targeting (IT) framework, the central bank commits to a publicly
announced numerical range for inflation, subordinates other intermediate targets, and
institutionalizes its commitment through a set of mechanisms that emphasize transpar­
ency and accountability for outcomes. Between 1989 and 2008, the number of central
banks practising full-fledged IT (FFIT) rose from zero to 27.^ South Africa became SSA s
first adopter in 2000, and in 2007 Ghana was the first low-income country in the world to
adopt the framework. Many other countries adopted elements of IT while retaining a
policy role for exchange rate or monetary targets; in SSA this list includes the region’s
two highest-performing economies, Botswana and Mauritius.
Ghana’s decision was consistent with a global pattern that has encouraged adoption of
FFIT at lower and lower levels of development. Durability has undoubtedly contributed:
as of 2009 no country had abandoned IT, despite the fact that very few adopters satisfied
all the preconditions emphasized in the early literature. But empirical studies also give
the framework high marks. On balance, the evidence suggests that inflation targeters
achieve more stable and somewhat lower inflation rates than non-targeters, with little or
no sacrifice in terms of employment and output; and that external shocks are somewhat
less likely to destabilize expectations (Mishkin and Schmidt-Hebbel 2007). The policy
literature has accommodated this ever more impressive track record by gradually
softening its concern for preconditions.^
In this chapter I review the logic of IT and examine its relevance for SSA. Issues like
fiscal dominance, supply shocks, and institutional development loom large in SSA,
relative to the constraints imposed—at least for now—^by high capital mobility. More­
over Africa’s central banks have already substantially increased the attention given to
inflation in the conduct of monetary policy, and with considerable success. The appeal of
FFIT may—ironically—^be strongest in the CFA zone, where it would replace the most
successful inflation anchor the continent has ever known.

' I use the IMF’s online listing of de facto monetary policy anchors in April 2008. The total is 28 if the
European Central Bank (which has officially resisted being characterized as an inflation targeter) is included.
^ Freedman, Laxton, and Otker-Robe (forthcoming) review IT in theory and practice.
360 MACROECONOMICS AND FINANCE

2. The global emergence of inflation targeting

The Great Inflation in the USA^—^with its global impact and long aftermath among
the emerging-market economies—created a revolution in the design and conduct of
monetary policy. By the late 1990s, economists and central bankers worldwide were
in agreement that good policy consisted in the exercise of ‘constrained discretion’
(Bernanke and Mishkin 1997). The intellectual underpinnings of this consensus are
well covered elsewhere (Woodford 2003). In practical terms, the Volcker disinflation of
1979-84 was a watershed. Technically the Fed did not become an inflation targeter in
1979. But the Fed publicly took on the inflation rate as the unambiguous responsibility of
monetary policy; it placed disinflation above other objectives; and it disinflated without
regard to the exchange rate. These themes were taken up worldwide over the following
two decades.
Attitudes towards the exchange rate evolved gradually, but by the late 1980s financial
liberalization among the industrial countries had unleashed short-term capital move­
ments on a global basis. Facing ever sharper dangers of speculative attacks and wishing
to liberalize their own financial controls, central banks began to move away from fixed-
but-adjustable pegs. Some adopted harder pegs, but the majority retained an active role
for monetary policy by choosing greater exchange rate flexibility.
FFIT therefore emerged in a context of global disinflation and amid a shift away from
exchange rate anchors. As nominal anchors, money growth targets were the obvious
alternative and were widely employed starting in the 1970s. But they were less transpar­
ent than exchange rates, and outside SSA they became associated with policy instability
under conditions of financial innovation. All anchors, moreover, were subject to credi­
bility problems. In 1989, the Government of New Zealand cut through these concerns
institutionally. Under the Reserve Bank of New Zealand Act, the central bank would be
legally bound to achieving a tight numerical range for inflation.^
The spread of FFIT after 1990 was influenced both by the success of adopters and by a
further development within academe. Taylor (1993) observed that Fed policy after 1980
was well approximated by a reaction function relating the policy interest rate to the
deviations of inflation from a target and output from its natural rate. The Taylor
rule gave operational content to constrained discretion, by reconciling a counter-cyclical
role for policy (discretion) with an aggressive response to inflation (constraint). By the
late 1990s it was clear that in models with forward-looking expectations and price
rigidities, Taylor-type rules not only anchored expectations but also generated highly
favourable combinations of volatility in inflation and output. The rule-based approach to

^ Roughly 1965-84; see Meltzer (2005).

■* Subsequent FFIT systems have been less draconian, but all constrain central banks to pursue numerical
targets for inflation.
 INFUTiON TARGETING 361

monetary policy evaluation bore more directly on constrained discretion than on FFIT
per se, but acquired an important role in the internal discourse of inflation-targeting
central banks (Berg, Karam, and Laxton 2006).

3. The African context

FFIT combines a novel choice of nominal anchor with a set of institutional commitments
designed to enhance the credibility and predictability of policy. How relevant are these
innovations for African countries?
The first of two dominant themes in Africa’s inflation history is the effectiveness of
monetary unions at maintaining hard exchange rate pegs and thereby anchoring the
inflation rates of member countries to those of the pivot currency (Masson and Pattillo
2005). The institutions of the CFA zone overcame fiscal dominance—a situation in
which public sector solvency has to be maintained through monetary finance, rather
than through adjustments in spending or taxes—by locating monetary policy at the
supranational level, where no single member government could dictate its terms. In the
institutional dimension, therefore, it is not clear that FFIT offers important advantages
for the fourteen CFA countries: the arrangements of the zone have been credible and
predictable.^ A hard peg, however, leaves very little room for discretion. If capital
mobility is high, it effectively gives up monetary policy altogether. This rigidity produced
a long and costly economic contraction in the zone when adverse shocks called for a real
depreciation in the 1980s and early 1990s and the relative price adjustment had to be
accomplished through deflation. A switch to inflation targeting, by allowing zone-wide
exchange rate flexibility, might produce a superior balance of constraint and discretion.
There is also a deeper institutional question about the appropriate locus of account­
ability. The convertibility guarantee from France and the EU underpins the viability of
the peg. While this form of dependence is almost certainly desirable relative to a
unilateral zone-wide peg, it may have long-run costs in terms of institutional develop­
ment, relative to a regime that can be managed autonomously by its own members.
The second dominant theme is that most of Africa did experience—and defeat—a
‘great inflation’ of its own. The phenomenon was restricted to the continent’s national
currencies, including the South African rand.® In contrast to the supranational institu­
tions of the CFA zone, national central banks lacked the legal or de facto autonomy to
resist fiscal pressures. By the early 1980s exchange rate anchors had been discredited—

^ Mali (rejoining in 1984), Equatorial Guinea (1985), and Guinea Bissau (1997) were willing to give up
their own currencies to join the zone.
^ The rand is the pivot of the Common Monetary Area, but Lesotho, Namibia, and Swaziland have no
influence over South Africa’s monetary policy.
362 MACROECONOMICS AND FINANCE

and marginalized—through a proliferation of distorting exchange controls and black

markets. The structural adjustment era swept away these controls, so that by the early
1990s central banks were committed to fostering macroeconomic stability indirectly,
through market-determined prices for foreign exchange and credit. Disinflation was by
no means immediate; letting exchange rates go unleashed inflation, particularly where
market-based reforms were undertaken before establishing fiscal discipline. The number
of countries with annual inflation above 40% peaked at eleven in 1995. But by the late
1990s, central banks across the continent had successfully employed money-based
disinflation programmes to bring inflation down into single or low double digits.
Tight fiscal rules (e.g. cash budgets) played a key role in overcoming fiscal dominance
during the 1990s. The majority of non-CFA central banks now operate reserve-money
frameworks that are the direct descendants of IMF conditionality programmes and
retain severe limits on central bank finance to government. These frameworks target
inflation in the well-defined sense that benchmarks for reserve-money growth are
derived from projections of inflation-consistent growth in money demand; in most
cases the central bank announces a desired range for inflation. Many countries now
also boast new legislative charters that clarify the central bank’s mandate and shelter it
from short-term political pressures.

4. Why add binding numerical targets?

Existing frameworks have performed well over the past decade, and have incorporated
many of the features of FFIT without committing to numerical targets. Outside the CFA
zone, most central banks are practising versions of what Stone (2003) calls ‘inflation
targeting lite’ (ITL), focusing seriously on inflation but also giving systematic attention to
exchange rate targets.^ Why go further? I offer observations rather than answers.
First, FFIT may ironically make more sense for African monetary unions than for
national currencies. For the CFA zone, FFIT combines a better choice of nominal anchor
with a transition path away from dependence on the exchange rate guarantee. A similar
argument holds for planned monetary unions elsewhere in Africa, where exchange rate
guarantees are not in prospect and the political instability of the major players—Nigeria
in West Africa and Kenya in East Africa—strengthens the case for union-wide restraints.
Second, FFIT may—ironically again—^be a more constructive choice for African
countries still struggling with fiscal dominance than it is for those who have overcome
it by other means. In contrast to the industrial-country situation, there is no empirical

^ On the prevalence of exchange rate objectives in SSA, see IMF (2008: Table 2.1).
 INFLATION TARGETING 363

consensus in favour of very low inflation in low-income countries.® This gives the welfare
logic of FFIT a distinctly second-best flavour in SSA: if credibility can be obtained
through other means, keeping inflation low and stable may not be the best way to use
it. However, where existing frameworks cannot be relied on to keep expected inflation
below (say) 15%, FFIT has a reasonable chance of targeting the policy distortion at its
source. The framework requires an explicit endorsement from the government, and it
directly educates the public that monetary policy cannot create high public-sector wages,
a competitive real exchange rate, or even cheap credit on a sustainable basis. This may be
the right path to credibility when external props are absent. South Africa provides an
example: without FFIT the Reserve Bank would undoubtedly be under even greater
pressure than it already is, following the populist transition within the ANC, to accom­
modate the food and fuel price shocks of 2007-8.^
Third, since FFIT was never designed for low-income countries, their economic
structure raises unresolved questions. Does it make sense to target inflation when
exchange rates, food prices, and public-sector prices (utilities, fuel, and public-sector
wages) are more prominent—and verifiable—in the public eye? Can the framework be
deployed successfully when supply shocks are dominant and the GDP gap is difficult to
measure? Is there a role for exchange rate targets, given imperfect capital mobility and
the importance of export promotion? What operational policy rules can be used when
there is not a strong transmission from policy interest rates to aggregate demand, and
where (as in almost all African cases) central banks use balance sheet instruments rather
than interest rates? Ghana faces all of these questions (Sowa and Abradu-Otoo 2009). As
of late 2009 the Bank of Ghana had yet to achieve a target range that was already higher
and wider than that of any other targeter. Its struggles with transparency are palpable; in
some quarters the Bank describes itself as practising ITL rather than FFIT (IMF 2008: 5).
It remains to be seen whether the distortions implied by committing to a target that is
rarely met, of unclear salience, and difficult to forecast are fundamentally damaging to
economic stability—or are more than compensated by committing the fiscal authorities
to self-restraint and forcing the pace of institutional maturation within the central bank.
Finally, recent events are testing inflation targeting in new ways. A confluence of food
and fuel price shocks in 2007 and 2008 pushed virtually all targeters above their ranges in
2008—the first time a miss of this scope had occurred. The global financial crisis sharply
reversed this impetus, confronting industrial-country targeters with risks of deflation,
and generating massive rescue operations with uncertain implications for future price
stability. These developments have dampened the short-run enthusiasm for FFIT in

® Modest inflation allows large relative price changes to occur without outright deflation. The CFA
experience of 1980-94 suggests, surprisingly, that deflation is damaging to growth even in low-income
economies with small formal sectors.
’ Frankel, Smit, and Sturzenegger (2007) discuss macroeconomic management in South Africa.
364 MACROECONOMICS AND FINANCE

Africa—fortuitously in my view, given the viability of existing ITL systems and the scope
for further improvements—while generating sharp reminders of the necessary regu­
latory complements to any successful monetary policy framework.

■ REFERENCES

Berg, Andrew, Philippe Karam, and Douglas Laxton, 2006. ‘A practical model-based approach to
monetary policy: overview’, IMF Working Paper 06/80, Washington, DC.
Bernanke, Ben S., and Frederic S. Mishkin, 1997. ‘Inflation targeting: a new framework for
monetary policy?’. Journal of Economic Perspectives 11.2, 97-116.
Frankel, Jeffrey, Ben Smit, and Federico Sturzenegger, 2007. ‘South Africa: macroeconomic
challenges after a decade of success’, RWP 07-121, John F. Kennedy School of Government,
Harvard University.
Freedman, Charles, Douglas Laxton, and Inci Otker-Robe, forthcoming. On Developing a Full-
Fledged Inflation Targeting Regime: Doing What You Say and Saying What You Do. Carleton
University and IMF.
International Monetary Fund, 2008. Ghana: Selected Issues. Country Report No. 08/332.
Washington, DC.
Masson, Paul, and Catherine PattiUo, 2005. The Monetary Geography of Africa. Washington, DC:
Brookings Institution.
Meltzer, Allan, 2005. ‘Origins of the Great Inflation’, Federal Reserve Bank of St. Louis, Review
87.2, part 2, 145-75.
Mishkin, Frederic S., and Klaus Schmidt-Hebbel, 2007. ‘Does inflation targeting make a differ­
ence?’, NBER Working Paper No. 12876, Cambridge, Mass.
Sowa, Nii Kwaku, and Philip Abradu-Otoo, 2009. ‘Inflation management and monetary policy
formation in Ghana’, in Gill Hammond, Ravi Kanbur, and Eswar Prasad (eds). Monetary
Policy Frameworks for Emerging Markets. Cheltenham: Edward Elgar.
Stone, Mark R., 2003. ‘Inflation targeting lite’, IMF Working Paper 03/12, Washington, DC.
Taylor, John B., 1993. ‘Discretion versus policy rules in practice’, Carnegie-Rochester Conference
Series in Public Policy 39, 195-214.
Woodford, Michael, 2003. Interest and Prices: Foundations of a Theory of Monetary Policy.
Princeton, NJ: Princeton University Press.