Abstract

After years of very high inflation prior to the mid-1990s, Mozambique has succeeded in stabilizing inflation, albeit with some lapses. A key factor in bringing inflation under control has been the implementation of a firm money-based stabilization program supported by prudent fiscal policies under successive IMF-supported arrangements. Financial liberalization and bank restructuring at early stages were also important in this regard. There have been, however, bouts of inflationary episodes driven by both domestic shocks (for example, droughts, floods, and banking crises) and external shocks (for example, inflows of foreign aid and fluctuations in oil prices). Thanks to monetary and financial sector reforms, Mozambique’s financial system, once government owned, has come a long way since the end of the civil war, evolving into a full-blown market-based system.

After years of very high inflation prior to the mid-1990s, Mozambique has succeeded in stabilizing inflation, albeit with some lapses. A key factor in bringing inflation under control has been the implementation of a firm money-based stabilization program supported by prudent fiscal policies under successive IMF-supported arrangements. Financial liberalization and bank restructuring at early stages were also important in this regard. There have been, however, bouts of inflationary episodes driven by both domestic shocks (for example, droughts, floods, and banking crises) and external shocks (for example, inflows of foreign aid and fluctuations in oil prices). Thanks to monetary and financial sector reforms, Mozambique’s financial system, once government owned, has come a long way since the end of the civil war, evolving into a full-blown market-based system.

The challenges for Mozambique—a post-stabilization economy—are to continue implementing an appropriate macroeconomic policy framework to consolidate macroeconomic stability, and to pursue a reform agenda aimed at deepening financial intermediation in a sound manner. The literature points to a number of areas that need to be considered. First, one must choose a consistent macroeconomic policy mix to consolidate macroeconomic stability (Sargent and Wallace, 1981; and Woodford, 2001), and a monetary policy framework to maintain inflation at single-digit levels (Fischer, 1993; Ghosh and Phillips, 1998; and IMF, 2005).1 Therefore, the next section begins by evaluating the role of macroeconomic stabilization policies and the monetary transmission mechanism in Mozambique, with the latter focusing on evaluating the appropriateness of the Bank of Mozambique’s (BM) current monetary targeting framework and mapping a way forward. Second, one must implement reforms that ensure the sound development of the financial sector. Financial development is recognized as important for economic growth (Levine, 2005), and it could also help promote price stability (Posen, 1995). Banking system distress and crises are also economically very costly (Frydl, 1999; and Noy, 2005). Thus, the section “Monetary and Financial Sector Reforms” examines Mozambique’s experience with these reforms as the country moves toward price and financial stability. The last section concludes by identifying the remaining challenges for Mozambique and lessons for other countries in sub-Saharan Africa at a similar stage of financial development and with a similar macroeconomic environment.

Stabilization Policies and the Monetary Transmission Mechanism

Macroeconomic Policy Mix

The effectiveness of monetary policy in controlling inflation depends critically on its coordination with fiscal policy. It is well established that governments running persistent deficits will, sooner or later, be forced to finance those deficits through money creation (seigniorage), thereby producing inflation (Sargent and Wallace, 1981). While this theory does not rule out the importance of other mechanisms that can fuel inflation and cause it to become persistent, fiscal imbalances have remained central to most models.2 The “fiscal view” of inflation has been especially prominent in the developing country literature, which has long recognized that less efficient tax collection, political instability, and more limited access to external borrowing tend to lower the relative cost of seigniorage and increase dependence on the inflation tax (Alesina and Drazen, 1991; Cukierman, Edwards, and Tabellini, 1992; and Calvo and Végh, 1999). As such, Catão and Terrones (2005) find that the higher the ratio of the fiscal deficit to narrow money, the higher the rate of inflation in developing countries, including those with moderate levels of inflation.3 Estimating ordinary least squares from the first quarter of 1996 to the third quarter of 2006, we find that, for an average M1/GDP ratio of about 7 percent in Mozambique, a 1 percent reduction (increase) in the ratio of the budget deficit to GDP lowers (raises) inflation by about 2½ percentage points on average, all else constant.4 This value lies at the low end of the range of the impact of deficits on inflation in low-income countries estimated in Catão and Terrones (2005) and is closer to the values reported for emerging markets. The result may reflect the short time span, which captures mostly the post-stabilization phase in Mozambique and the absence of nontraditional channels of fiscal influence on inflation—namely monetization expectations and the wealth effects of public debt—which can be formed independently of the size of the budget deficit (Kwon, McFarlane, and Robinson, 2006).

To account for the different channels of fiscal policy effects on inflation, one could focus on the role of public debt—instead of the fiscal deficit—in determining inflation and inflation expectations. Using a simplified forward-looking model of inflation following Castro, De Resende, and Ruge-Murcia (2003), Kwon, McFarlane, and Robinson (2006) establish a linear relationship between inflation and increases in money supply and public debt (both domestic and external) so that the relationship can be tested empirically. Given the limited time-series data for individual countries, Kwon, McFarlane, and Robinson (2006) employ panel data techniques to allow for the variability of individual countries while preserving the dynamics of adjustment within countries. The regression results show that debt growth has a strong and stable positive effect on inflation in low-income economies and the smaller advanced economies.5 The coefficient for public debt is nearly 0.2 for the short term, implying that a 1 percent increase in public debt leads to a 0.2 percentage point increase in inflation.

The coefficients are lower than those of money growth but are significant at the 5 percent level. The strong debt-inflation linkage, after controlling for money growth, is consistent with the significant positive relationship between fiscal deficits and inflation only during high-inflation episodes in low-income countries (Catão and Terrones, 2005; and Fischer, Sahay, and Végh, 2002). Kwon, McFarlane, and Robinson (2006) also undertook a simple vector autoregression (VAR) to trace the transmission channels of the fiscal influence on inflation. The panel VAR outcomes lend additional support to the prediction of the fiscal-monetary model of inflation—that the debt-inflation link is affected by institutional and structural factors. Impulse responses for low-income countries show a strong and positive response of money supply and inflation to fiscal shocks, whereas the impulse responses for major advanced economies do not. This suggests that increases in public debt in developing countries are more often than not accommodated by monetary easing—a phenomenon known as fiscal dominance.

We apply a VAR to Mozambique as in Kwon, McFarlane, and Robinson (2006) to identify the transmission channels of the fiscal influence on inflation and to test whether the cross-country debt-inflation relationship identified from the panel regressions holds for Mozambique. Our VAR consists of inflation and growth of public debt (both domestic and external), money, and real GDP from the first quarter of 1996 to the fourth quarter of 2006. Impulse responses are based on the Cholesky decomposition of the structural shocks in the order of output, public debt, money, and prices (see Figure 4.1).6 The choice of lag length is determined by the Schwarz criterion and the Akaike information criterion. The VAR outcomes confirm the significance of public debt dynamics in determining inflation in Mozambique. The impulse response functions show that the price level is positively affected by money supply and public debt but that the latter has a longer-lasting effect on inflation. The variance decompositions suggest that money plays a greater role in explaining inflation dynamics over the first two quarters but that fiscal shocks start to play a similar role thereafter (see Figure 4.2). Also, expansionary fiscal shocks have expansionary effects on money supply while the opposite does not hold. These results are similar to those from the panel VAR estimates for low-income countries in Kwon, McFarlane, and Robinson (2006) and robust to changes in the ordering of the shocks. The impact of only domestic debt shocks on headline inflation and base money follows a similar pattern (and is thus not reported here) but is stronger than total public debt, highlighting the important role played by net domestic credit to the government.

Figure 4.1.
Figure 4.1.

Public Debt, Money, and Inflation

Source: Authors’ calculations.
Figure 4.2.
Figure 4.2.

Public Debt, Money, and Inflation

Source: Authors’ calculations.

Caution is needed, however, in interpreting these results. The implications of rising public debt for inflation are observationally similar in the Sargent-Wallace framework (1981) and the Fiscal Theory of the Price Level (FTPL). Nonetheless, there is an important theoretical distinction between the two (Leeper and Yun, 2005). Under the FTPL, an increase in public debt raises the wealth of the holders of government bonds without reducing the wealth of others. As bond prices rise, aggregate demand for goods and services increases, pushing up prices in the general economy. Under the Sargent-Wallace framework of the so-called unpleasant monetarist arithmetic, an increase in government debt that is not fully backed by a future real primary surplus will increase concerns about the monetization of public debt and raise inflation expectations, thereby reducing demand for government bonds and boosting interest rates. With both models predicting higher prices in response to rising public debt, and underdeveloped bond markets providing little information on market-determined interest rates, it is difficult to infer which is the dominant force in Mozambique. Notwithstanding, the VARs confirm that movements of public debt (particularly domestic debt) do matter for inflation dynamics in Mozambique, providing scope for fiscal consolidation and reduction of public debt, including limiting net issuance of domestic debt for sterilization purposes to further support a disinflation program.

By the same token, the credibility of a given fiscal path is also important. For example, in a highly aid- and resource-dependent economy like Mozambique’s, in the absence of institutional mechanisms that ensure rapid fiscal adjustment in the event of declines in foreign aid (or the country’s inability to lock in high aid levels through firm medium-term commitments by donors) or resource revenues, it may be impossible to rule out the expectation of an increase in future required seigniorage. In such a situation, the literature indicates that fiscal rules that limit the size of budget deficits or public debt could, under appropriate circumstances, be an important institutional means of safeguarding price stability (see Chapter 6). The independence of the central bank could also help reduce monetization concerns, and the development of the financial sector could help promote price stability, as the financial sector tends to support the central bank’s policy autonomy (Posen, 1995).

Monetary Transmission Mechanism and Appropriate Policy Framework

The easing of concerns regarding fiscal dominance and the removal of distortionary controls over domestic interest rates and liberalization of the financial sector provide scope for a more activist role for monetary policy.7 The domestic financing requirement of the central government budget in Mozambique has been falling and negative in recent years, particularly credit from the central bank.8 As indicated above, this helped Mozambique reduce inflation to single digits and is likely to have mitigated concerns about the monetization of public debt.9 In fact, the medium-term fiscal framework, which was approved for the first time by the Council of Ministers in 2006, explicitly avoids recourse to domestic financing. In such an environment, many view the current monetary policy setting in sub-Saharan Africa as an interim stage in a move toward wider adoption of formal inflation targeting practices in which inflation (more precisely, expected inflation) is the intermediate target, instead of either some monetary aggregate or the exchange rate, and where the interest rate rather than base money is the operational instrument (see Adam and O’Connell, 2005).10 Thus, while elements of this debate will necessarily reflect themes in the current literature on monetary policy in emerging market countries (for example IMF, 2005; and Hakura, 2005), to the extent that post-stabilization economies like Mozambique are free of fiscal dominance, current policy arrangements are inflation targeting frameworks in the broad sense of being centered on maintenance of a nominal anchor (see Adam and O’Connell, 2005). In this respect, therefore, the relevant policy questions are not only how, and over what horizon, countries may make the move toward formal inflation targeting but also whether and how the operation of monetary policy can be improved given current monetary frameworks.11 To consider this issue, we turn to an empirical evaluation of the monetary transmission mechanism, followed by a look at the monetary and financial sector reform process in Mozambique.

There has been a burgeoning literature on the transmission mechanism of monetary policy in many countries. Typically, this strand of research has been conducted in the context of a VAR framework pioneered by Sims (1980). Notable examples using VAR to identify transmission of monetary policy for advanced economies include Christiano, Eichenbaum, and Evans (2000) for the United States; Kim and Roubini (2000) for the G-7 economies; and Peersman and Smets (2003) for the euro area. A few recent papers have also studied the monetary transmission mechanism for a number of countries in sub-Saharan Africa. For example, Saxegaard (2006) uses a threshold VAR model for a number of sub-Saharan African countries and finds that excess liquidity in the region weakens the monetary transmission mechanism and, thus, the ability of the monetary authorities to influence demand conditions. Aron and Muellbauer (2001) analyze different aspects of the transmission mechanism in South Africa, including the degree of exchange rate pass-through and the impact of openness. However, there has been little analysis of the monetary transmission mechanism in Mozambique and other post-stabilization economies in sub-Saharan Africa, leaving a number of unanswered questions and somewhat of a vacuum of knowledge in this area. We will therefore examine the monetary transmission mechanism in Mozambique and explore the following questions:

  • What role do monetary aggregates (and credit) and interest rate shocks play in transmitting monetary impulses through the economy?

  • What is the degree of exchange rate pass-through?

  • Are there measures of core inflation that are better controlled by monetary policy?

Structural VAR Modeling

Following Kim and Roubini (2000) and Sims and Zha (2006), we assume the economy is described by a structural-form equation:

G(L)Yt=C(L)χt+ɛt,

where G(L) is an n × n matrix polynomial in the lag operator; C(L) is an n × k matrix polynomial in the lag operator; Yt is an n × 1 vector of endogenous Mozambican variables; and χ is a k × 1 vector of exogenous foreign variables; ɛt is an n × 1 vector of structural disturbances, with var(ɛt) = Λ, where Λ is a diagonal matrix and the diagonal elements are the variances of structural disturbances; therefore, structural disturbances are assumed to be mutually uncorrelated.

Corresponding with this structural model we can estimate a reduced-form VAR:

Yt=A(L)Yt+B(L)χt+μt,

where A(L) and B(L) are matrices polynomials; μt is a vector of reduced-form disturbances, with var(μt) = Σ.

There are many ways of recovering the parameters in the structural-form equations from the estimated parameters in the reduced-form equation. Some methods give restrictions on only contemporaneous structural parameters. A popular and convenient method is to orthogonalize reduced-form disturbances by Cholesky decomposition (as in Sims, 1980, among others). However, in this approach to identification, we can assume only a recursive structure. Sims (1986) suggests a generalized method (structural VAR) in which nonrecursive structures are allowed while still giving restrictions only on contemporaneous structural parameters.

We assume the exogenous vector χt contains a terms of trade index for Mozambique (TOT), foreign aid inflows (AID), and the U.S. federal funds rate (USR):

χt=[TOT  AIDUSR]

These variables are included to control for changes in the overall global economic stance (USR) and fluctuations in the prices of Mozambique’s main exports and imports (TOT). We also include foreign aid (AID) since it is found to improve the fit of the model—not surprisingly, given the magnitude of aid in Mozambique, which is consistently above 10 percent of GDP.

Since the monetary authority follows a feedback rule—it sets monetary policy in response to economic news—it is important to control for the systematic component of the policy rule in order to identify exogenous monetary policy changes. If the monetary authority tightens monetary policy in response to a negative and inflationary supply shock, the ensuing recession and price inflation are due not only to the monetary contraction but also to the original negative supply shocks. To identify the part that is due to monetary policy alone, we include the price of oil (GAS), a proxy for negative and inflationary supply shocks. The other endogenous variables include real GDP, the headline consumer price index (CPI), the domestic treasury bill interest rate (R), the base money stock (M), and the nominal exchange rate against the U.S. dollar (E):

Yt=[GAS  GDP  CPIRME].

The system encompasses the key sources of inflation identified by Loungani and Swagel (2001) and Barnichon and Peiris (2007) in their analysis of the sources of inflation in developing countries, particularly in Africa. First, as discussed by Agénor and Montiel (1999), inflation in developing countries is often linked to underlying fiscal imbalances. Such imbalances can lead to an increase in inflation by causing excessive money creation, as in Sargent and Wallace (1981), or by triggering a balance of payments crisis and resulting in an exchange rate depreciation, as in Liviatan and Piterman (1986). Another source of inflation may be due to macroeconomic overheating—that is, an excessive expansion of aggregate demand over potential output supply—as examined by Coe and McDermott (1997) for 13 Asian economies, estimated by the influence of an activity variable such as GDP. A third source of inflation, examined by Ball and Mankiw (1995), is supply-side “cost shocks”—movements in the prices of particular goods, such as oil, that lead to persistent changes in the aggregate price level. Finally, as discussed by Chopra (1985), inflation may have a substantial inertial component arising from the sluggish adjustment of inflationary expectations or the existence of staggered wage contracts (see Calvo, 1983; and Christiano, Eichenbaum, and Evans, 2001). There may also be inflationary shocks coming through shocks to marginal costs as postulated by the more recent literature on New Keynesian Phillips Curves (see Galí and Gertler, 1999).

Estimations based on other interest rates (rediscount rate); monetary aggregates (M2, M3, and domestic credit to the private sector); exchange rates (nominal effective exchange rate and exchange rate against the South African rand); real activity measures (exports); and consumer price indices (for nonfood items and energy) were also examined but are not presented here as the results were largely similar to the baseline model.

Identification scheme 1: Recursive VAR

The first identification scheme is the standard approach that imposes a recursive structure of the VAR, with the ordering of variables given by Yt. Intuitively, it assumes that GDP has no contemporaneous effect on oil prices (GAS); prices (CPI) have no immediate effect on output (GDP); interest rates (R) have no immediate effect on prices; monetary aggregates (M) have no immediate effect on interest rates; and the nominal exchange rate (E) has no immediate effect on monetary policy. This amounts to estimating the reduced form, then computing the Cholesky factorization of the reduced-form VAR covariance matrix.

Identification scheme 2: Non-recursive structural VAR

Unlike the recursive identification, we assume a contemporaneous relation between interest rates, money, and the exchange rate, following Kim and Roubini (2000) and Sims and Zha (2006). Interest rates are allowed to respond contemporaneously to shocks to output, prices, and the exchange rate. Money responds to interest rates and the exchange rate but does not respond immediately to contemporaneous output and price shocks. Following previous work by others (for example, Sims, 1986), we use as an identifying restriction the idea that monetary policy cannot respond contemporaneously to disturbances in the general price level or the level of GDP. The argument for this restriction is based on the absence of contemporary data on these variables at the time policy decisions have to be made. This assumption gains some credibility in the current paper relative to previous applications of it, however, because we allow the contemporaneous response of monetary policy to global oil prices. The nominal exchange rate responds immediately to all other variables.

The VAR models are estimated using quarterly data between 1996 and 2006.12 All variables are in log-differences, except interest rates, which are in percentage terms.13 This is likely to result in the loss of information on long-run relationships between the variables in the system, which is a weakness in the approach. However, given the short time span of the sample, we do not consider that an explicit analysis of the long-run behavior of the economy would have been fruitful or essential to answering the questions at hand.14 Standard information criteria are used to select the lag lengths of the VARs, which turn out to be two quarters.

Results

The results of the baseline model under the two identification schemes were broadly similar; therefore, only the impulse responses and variance decomposition of the non-recursive structural VAR are shown in Figure 4.3. The first set of graphs in Figure 4.3 display the impact (the impulse response) of a one standard deviation oil price, GDP, headline inflation, interest rate, monetary policy (base money), and exchange rate shock (defined as an exogenous, unexpected, temporary rise at t = 0) on GDP, headline inflation, and the nominal exchange rate. The relative importance of the exogenous and monetary policy shocks for fluctuations in output, prices, and the nominal exchange rate at different forecast horizons can be gauged through the forecast error variance decompositions.

Figure 4.3.
Figure 4.3.
Figure 4.3.
Figure 4.3.

Effects of Exogenous and Monetary Shocks (Recursive VAR), 1996–2006

The key insights are as follows:

  • The effect of monetary impulses on GDP appears to be insignificant, while oil price and exchange rate shocks possibly have a modest impact on the real economy. This could be due, in part, to measurement problems with the constructed GDP series, although the impact on alternative real activity variables such as exports and non-megaproject exports also does not seem significant; the impact of exchange rate shocks on non-megaproject exports is more discernible.

  • A monetary policy shock has a significant impact on prices. For example, the percentage change in the price level over an initial percentage change in base money is 0.21 for headline inflation and 0.12 for core (nonfood and energy) inflation after two quarters. The pass-through of M2 and M3 is also significant, albeit less pronounced, while the impact of interest rate shocks on prices is insignificant. Monetary aggregates do not have a more discernible or larger impact on measures of core inflation. There is also no evidence of a significant credit channel (shocks to domestic credit to the private sector) to prices.

  • The exchange rate pass-through to headline inflation is greater than measures of core inflation.15 The cumulative impulse response to an exchange rate shock of one standard deviation (or 8 percent) gives an exchange rate pass-through elasticity (percentage change in price level t periods after the shock over an initial percentage change in the exchange rate) of 0.36 for headline inflation after four quarters. The pass-through to core inflation is negligible. This suggests that the exchange rate pass-through to headline inflation takes place almost exclusively through food prices. The less than one-to-one exchange rate pass-through may be explained by the degree of openness and strategic behavior in the form of pricing-to-market under imperfect competition.

  • The nominal exchange rate does not appear to be susceptible to monetary and interest rate shocks. The latter suggests a somewhat low degree of capital mobility in Mozambique, although one should not be too surprised to find it hard to explain nominal exchange rate movements (see Engel, Nelson, and West, 2007).16

Economic Interpretations of the Econometric Results

The empirical findings suggest that headline inflation in Mozambique is significantly influenced by monetary policy (base money). Therefore, the results confirm that firm base money control has played an important part in bringing inflation under control. But they also show that times of lax monetary policy owing to weaknesses in the monetary and liquidity management framework and to events such as the banking crises that undermined monetary control have contributed to greater price instability (see “Monetary and Financial Sector Reforms” below). In addition, headline inflation is explained largely by its own innovations and, with food items weighted heavily in the CPI basket, is no doubt susceptible to wild swings in food prices caused by weather-related shocks to agriculture. Moreover, monetary aggregates and interest rates seem to have little impact on real output. This may be the result of structural weaknesses in the financial sector, which, at least until recently, are likely to have hampered the transmission mechanism of monetary policy. These weaknesses are likely to have made it difficult to cushion exogenous shocks in a systematic and credible way, on occasion causing substantial volatility in both prices and nominal interest rates. Volatility was also probably further exacerbated by Mozambique’s shallow foreign exchange and public debt markets, with exchange rates and interest rates responding to numerous shocks not under the control of the central bank.

The base money–targeting approach seems to have been an appropriate policy framework, and the existence of the target appears to have anchored inflationary expectations and provided some discipline to discourage overly inflationary policies, albeit with some lapses. The BM has been targeting the monetary base—currency and bank reserves—as a way to achieve its longer-term goal of price stability. For an operational target to succeed in achieving a price objective there should be an identifiable transmission mechanism from the target to the price objective so that policymakers can determine the level they should seek. Our results show that base money does well on this count, similar to the findings in Barnichon and Peiris (2007), where the money “gap” appeared to contain significant information regarding the evolution of inflation in sub- Saharan Africa, although the transmission of monetary impulses through the real economy—and particularly to output—is less clear. Also, interest rates do not have an identifiable transmission mechanism to prices, making them unsuitable as instruments of monetary policy for now. In addition, policymakers need an active market in the target instrument that is not dominated by the central bank. In Mozambique, secondary debt markets have limited activity, with no overnight trades at all on some days. Targeting a rate in a market that barely trades would be difficult or impossible. Even if enough trading developed to make such targeting mechanically possible, it is not clear that achieving the targeted interest rate would result in the desired outcomes for the broader market structure or the economy.

Looking forward, the pursuit of a “lite” inflation targeting regime (Stone, 2003) could be seen as a transition to a full-fledged inflation targeting regime. The BM conducts monetary policy in a difficult setting characterized by a high propensity to transact in cash, shallow domestic financial markets, and an economy prone to numerous exogenous shocks. This is further complicated by large foreign exchange inflows in the form of grants and loans to finance the country’s large fiscal deficit and poverty reduction strategy, which injects vast amounts of liquidity into a thinly monetized economy. Against this background, Chapter 5 shows that a judicious mix of foreign exchange sales and domestic debt issuance to mop up excess liquidity and target inflation could help the monetary authorities maintain a low level of macroeconomic volatility. Such a regime, which employs less market-oriented monetary targets, is close to the current base money–targeting regime in terms of its operational framework, can be implemented gradually, and can serve as an intermediate step toward a full-fledged inflation targeting regime.17 In the next few years, by putting in place some of the preconditions for implementing a full-fledged inflation targeting regime (see Carare and others, 2002, and Box 4.4), Mozambique may gradually transition to such a regime.

Monetary and Financial Sector Reforms

First Wave of Reforms (1992–2000)

Liberalization of the financial system was one of the main objectives of Mozambique’s successful post-conflict reform program during the 1990s and was an important element in IMF-supported programs during the period.18 In 1993, the BM moved to a market-determined exchange rate system and established an interbank foreign exchange market. Interest rates were fully liberalized in 1994, but real interest rates were not consistently positive until 1997 owing to structural obstacles, including state ownership of the two dominant banks. Monetary management gradually evolved from a reliance on credit ceilings on individual banks to the use of reserve requirements and, starting in 1997, indirect instruments of monetary control. These measures have enhanced monetary control, helping to bring inflation down from very high levels and substantially decrease its variance (Figure 4.4). In line with economic theory and previous country experiences (McKinnon, 1973; Shaw, 1973; and Mirakhor, 1990), interest rate liberalization has also jump-started financial savings (Figure 4.5).

Figure 4.4.
Figure 4.4.

Mozambique: First-Generation Reforms and Inflation Dynamics, 1996–2000

(In percent)

Source: IMF, International Financial Statistics.
Figure 4.5.
Figure 4.5.

Mozambique: First-Generation Reforms and Financial Intermediation

(In percent)

Source: IMF, International Financial Statistics.

Financial sector reforms during this period also focused on developing the legal framework and establishing a basic supervisory regime for a more competitive market-based banking system. Several theoretical and empirical studies have shown that increasing banking competition by reducing government ownership as well as by removing obstacles to foreign entry are an important trigger to the subsequent development of sound, deep, and more efficient financial systems (Shleifer and Vishny, 2002; La Porta, Lopez-de-Silanes, and Shleifer, 2002; Barth, Caprio, and Levine, 2004; and Claessens and Laeven, 2004).

Mozambique’s reform experience appears to corroborate this hypothesis while emphasizing the importance of complementing it with a well-managed privatization process and appropriate regulatory and supervisory practices. The Financial Institutions Law and the Central Bank Law were passed in 1991, introducing banking supervision rules that were later supplemented by prudential regulations and the introduction of on-site and off-site inspections. Institutional reforms focused on separating the commercial and central banking functions of the BM. These reforms were accompanied by the restructuring and cleanup of the commercial arm of the BM, which was renamed Banco Comercial de Moçambique (BCM) in 1992. BCM was partially privatized, along with Mozambique’s other main government-owned bank, Banco Popular de Desenvolvimento (BPD, renamed Banco Austral, BA), between 1996 and 1997. Obstacles to private and foreign ownership of banks were removed, leading to a marked increase in the number of banking institutions, most of them foreign owned. On balance, foreign bank entry has brought Mozambique the knowledge and skills required to establish a proper market-based financial system, as well as additional resources for financial intermediation. The partial sale of BPD and BCM to a foreign investor of dubious quality, on the other hand, demonstrates the shortcomings of a poorly managed privatization compounded by weak supervisory practices, as described below.

Remaining Weaknesses in the Monetary and Financial System in 2002

Notwithstanding initial advances brought by the first wave of financial sector reforms, a number of structural problems with a direct impact on the stability, soundness, depth, and efficiency of the financial system remained.

Despite improvements in monetary management, inflation rates remained relatively high and volatile (Figure 4.4). Although this outcome was due, in part, to Mozambique’s vulnerability to significant supply shocks (for example, related to weather and terms of trade), the volatility of prices and nominal interest rates highlighted weaknesses in the monetary framework and public debt management (Box 4.1). Such weaknesses prevented monetary authorities from cushioning largely fiscally induced demand shocks (that is, the unanticipated influx of external grants and loans, and costly recapitalizations described above) in a systematic, credible, and transparent way, causing, at times, substantial volatility in prices, as well as in nominal interest and exchange rates. Shallow foreign exchange and public debt markets exacerbated this volatility, prompting the government to adopt temporary measures to tightly manage interest and exchange rates—an action that reversed some of the original gains from financial liberalization and further compromised the transparency of the monetary framework.

Weaknesses in Monetary Policy Management

Before Mozambique’s banking crisis in 2000–02, effective monetary policy and public debt management in Mozambique were compromised by weaknesses in the monetary framework and in the Bank of Mozambique’s (BM) net financial position.

The monetary framework lacked transparency. Market participants did not fully understand the goals or operational framework of the country’s monetary policy, in part because of insufficient efforts by the BM to communicate with market participants, and also because of inconsistencies in the use of instruments that sent conflicting signals on the monetary policy stance.

BM’s fragile financial position put the central bank’s operational independence at risk. While the BM reported positive capital and declared operating profits, its net worth and profits became negative once valuation adjustments were properly accounted for. The BM’s losses originated mainly from (1) accumulated foreign exchange valuation losses on its external liabilities to foreign lenders, and (2) operating losses stemming from sterilization (costs of monetary policy) not transferred to the budget.

Although a basic supervisory framework for the banking system had been put in place, the government did not have the capacity to effectively implement the new regulations; this problem was compounded by the absence of adequate reporting standards for monitoring the soundness of bank lending. The soundness of the financial system was further undermined by an increasing degree of liability dollarization, with foreign currency–denominated loans accounting for 70 percent of total loans in 2002. Moreover, despite their partial privatization, the two main state-owned banks remained subject to political interference, and a weak legal environment for lending (Box 4.2) translated into a rising level of nonperforming loans (especially for BCM and BA), high lending rates, and a poor level of financial intermediation even by regional standards, as reflected in Mozambique’s low credit-to-GDP and loan-to-deposit ratios (Table 4.1).

Table 4.1.

Mozambique and Comparators: Financial Intermediation, 2002

article image
Sources: IMF, International Financial Statistics; and Barth, Caprio, and Levine (2004).

Mozambique’s high lending rates and poor financial intermediation were a result of numerous structural factors. Real lending rates in meticais during 1999–2002 averaged 17.5 percent. Such high and rigid rates were due primarily to very large intermediation spreads, which, according to a decomposition exercise conducted in Mozambique’s 2003 Financial Sector Assessment Program (FSAP)19 (Figure 4.6), were caused partly by high levels of nonperforming loans made to state-owned enterprises and other noncompliant borrowers in the aftermath of the bank privatization process in the mid-1990s and banking crisis of 2000–02. Other factors included (1) high overhead costs that reflected, in part, the small size of the Mozambican financial system and credit risks; (2) high reserve requirements; and (3) generally substantial profit margins, reflecting high bank concentration and limited competition.

Figure 4.6.
Figure 4.6.

Mozambique: Decomposition of Interest Rate Spreads in Meticais

Source: Bank of Mozambique.Note: 2002 data.

Weaknesses in the Institutional Lending Environment

Mozambique’s high credit risk was attributed partly to its lack of institutional capacity to help lenders make informed decisions by disseminating information on borrowers’ creditworthiness or to enable lenders to collect from borrowers upon default. Debt enforcement for unsecured loans through the courts was inefficient, and the country lacked both comprehensive laws for enforcing the collection of the collateral for secured loans and efficient bankruptcy laws.

uch04fig01

Days to Resolve Business Dispute, 2002

Source: World Bank (2003).

In the absence of institutions that could facilitate loan recovery and contract enforcement, banks were compelled to adopt a conservative stance toward smaller borrowers, restricting access to credit and charging higher rates. Lack of information on borrowers’ creditworthiness further aggravated this problem; indeed, it increased uncertainty regarding nonperforming loans, forcing banks to increase their loan loss provisions (which made credit more costly), impose stringent collateral requirements, or simply refuse credit to applicants with no previous credit history.

Weaknesses in Mozambique’s financial system became particularly apparent during the costly recapitalization operations that took place in the aftermath of the bank privatization process. The privatized banks performed poorly because of continued insider lending and lending to parastatals, and the weak regulatory environment. As a result BA became insolvent. It was renationalized in 2000 and then reprivatized when it was sold to Amalgamated Banks of South Africa (ABSA) in 2001. ABSA accepted only 20 percent of BA’s loans, so the Mozambican government had to provide bonds to cover the remaining 80 percent at a cost of approximately US$107 million. Likewise, BCM also continued to report losses after its privatization and was recapitalized by its shareholders in late 2000 and early 2001 at a total cost of US$130 million, half of which was paid by the government. This amount turned out to be insufficient, and an additional bailout of US$11 million (of which US$2.5 million came from the government) was necessary once the merger of BCM with Banco Internacional de Moçambique turned the new bank (currently referred to as BIM) into a market leader considered “too big to fail.”

Lessons from the 2000–02 Banking Crisis and Subsequent Reform Efforts

Benefiting from recommendations from Mozambique’s 2003 FSAP and with technical assistance from the IMF, the World Bank, and other donors under the program known as the Financial Sector Technical Assistance Project (FSTAP), the Mozambican government initiated well-sequenced reforms aimed at strengthening banking supervision, prudential regulations, and financial reporting in line with international best practices.20 To limit the volatility in prices and interest rates, it made refining monetary policy formulation and deepening monetary and public debt markets top priorities. The government also initiated a more gradual program of reforms directed toward improving the legal and institutional environment for lending and then toward promoting the sound development of non-bank financial institutions (NBFIs), initially by enhancing the regulatory and supervisory framework.21

Strengthening Banking Supervision and the Regulatory Environment

Measures adopted over the past four years included the following:

  1. Comprehensive diagnostic reviews of the major banks by international firms according to International Financial Reporting Standards (completed in 2005).

  2. The implementation in 2004 of the new Financial Institutions Law, strengthening the central bank’s supervisory powers and introducing stricter penalties for noncompliance.

  3. Strengthening banking supervision in line with the Basel Core Principles by building on- and off-site supervisory capacity, requiring financial institutions and corporations to prepare financial statements in compliance with international accounting standards, and adopting prudential regulations that reflect international best practices.

  4. Adoption of International Financial Reporting Standards by the BM in 2006 and by commercial banks in 2007.

Mozambique’s reform effort has been noticeable. Over 2004–06, the government demonstrated its commitment to returning banks to private ownership; it has announced its intention of withdrawing from the one bank in which it remains a minority shareholder. Reform efforts have been pronounced in the area of banking supervision and reporting standards (Box 4.3). A new law strengthening the BM’s supervisory and enforcement mandates was implemented in 2004. To dispel uncertainties about the overall condition of the banking system following the restructuring, the government allowed international firms to carry out comprehensive diagnostic reviews of the major banks according to international financial reporting standards (IFRS). These reviews, which were concluded in 2005, endorsed the overall soundness of the banking system. The authorities also continue to make progress in aligning on- and off-site banking supervision practices and accounting standards with international best principles.

While there has been some recent progress in improving the monetary policy framework, important measures to enhance monetary policy operations and public debt management have yet to be implemented. The BM has already taken important steps to strengthen its monetary policy operations by increasing the use of foreign exchange sales and issuance of treasury bills to sterilize excessive liquidity, and it is currently working to improve liquidity management, ensure its operational independence, and deepen foreign exchange, money, and debt markets (Box 4.4). Over time, these measures should set the stage for the BM to gradually transition to a “lite,” and subsequently to a full-fledged, inflation targeting regime.

Reforms in the institutional lending environment have followed but have lagged behind other reforms. Among the few achievements on the legal front have been the implementation of the new Commercial Code and the establishment of the commercial sections of the judicial tribunals in the cities of Maputo, Beira, and Nampula. The Commercial Code will help relieve the overburdened judicial system by making commercial disputes less likely. The commercial sections of the judicial tribunals became operational in 2007, and are expected to accelerate the recovery of unsecured loans once they have received appropriate equipment, facilities, and technical assistance. Debt recovery is expected to be accelerated even further once a revised Civil Procedure Code with streamlined judicial procedures is enacted. However, despite these encouraging steps, significant measures are still required to improve the lending environment (Box 4.5).

The government has taken initial steps to support a sound expansion of NBFIs. The BM has started to license and supervise deposit-taking MFIs. A new law on social protection recently approved by the National Assembly will help strengthen the social security and supplementary pension systems. As part of the restructuring of the INSS, an actuarial study to determine whether the current level of contributions is capable of sustaining the benefits covered will be completed in 2008. The government is also enhancing the regulatory and supervisory framework for the pension and insurance sectors. The Insurance Law is being reviewed in line with international best practices, and revised prudential and solvency requirements for Mozambican insurers are being developed, while a new IFRS-compliant Chart of Accounts for insurers will also be designed. New regulations are also being prepared for supplementary pension funds. Looking ahead, the focus will need to shift to strengthening the capacity of industry regulators.

Gauging the Impact of Reform

Financial sector reforms, while still incomplete, have contributed to marked improvements in the soundness and efficiency of the banking system. Bank profitability has recovered, and the share of nonperforming loans has fallen to less than 5 percent of all loans since 2005, according to Mozambican accounting standards (Figure 4.7). Greater confidence in the banking system has led to a significant decrease in liability dollarization and thus less financial vulnerability.22 Banks have also become more efficient. In particular, the ratio of noninterest expenses to gross income and, to a lesser extent, the ratio of personal expenses to noninterest expenses have been declining since 2004.

Figure 4.7.
Figure 4.7.

Mozambique: Banking Soundness, Depth, and Efficiency, 2000–06

Source: Bank of Mozambique.1 According to Mozambican accounting standards, nonperforming loans include only part of past-due loans.

Strengthening Monetary and Public Debt Management

Monetary policy framework. The BM has adopted a money targeting framework using base money as the operational target. In order to improve the BM’s transparency and communication vis-à-vis market participants, the BM has approved a long-term monetary policy strategy document clarifying the money targeting framework, defining a new format for the monetary policy committee, and specifying the BM’s communication policies. The central bank has also made good progress in increasing its analytical capacity through the development of databases and the use of standard monetary models in assessing the appropriateness of the monetary policy stance. Work is ongoing to translate the annual and quarterly monetary targets into a liquidity management plan that separates the management of short-term “temporary” and long-term “structural” liquidity resulting from government domestic expenditures financed mainly through foreign donor inflows. The BM introduced repos in 2007 to fine-tune temporary liquidity variations and is working to ensure an appropriate mix of foreign exchange sales and net treasury bill issuance for the purpose of sterilizing structural liquidity so as to facilitate absorption of foreign aid, with clear signaling of sterilization and intervention actions in the foreign exchange market. The deepening of debt and foreign exchange markets will also be needed to avoid undue movement in interest rates and exchange rates.

Central bank independence. The central bank’s operational independence in conducting sterilization operations has been secured through a more comprehensive agreement to shift the corresponding costs to the Ministry of Finance. Further reforms are also needed to clarify that the central bank’s sole objective is price and financial stability while legally ruling out the possibility of financing the fiscal deficit, to further enhance central bank autonomy and credibility.

Central bank liquidity management. The BM is in the process of refining its liquidity monitoring and forecasting with technical assistance from the IMF. Additional technical assistance is needed to enhance the BM’s capacity to accurately forecast short-term liquidity flows in the financial system and manage temporary fluctuations in liquidity using appropriate instruments. Of critical importance are improvements in the information provided by the Treasury on its daily cash flows and use of repos to fine-tune temporary liquidity variations (see previous page).

Public debt management. Reform is still in its early stages. The government needs to develop a strategy for the management of domestic debt, and the public debt department at the Ministry of Finance will need to be restructured and its staff trained.

Treasury bill and money markets. While recent progress was achieved with the removal of interest caps on T-bill auctions, more needs to be done to promote an active and liquid market for government securities. At the top of the list would be improving the enforcement of the legal framework governing the transfer of securities and the implementation of a real-time gross settlement system to foster the development of a secondary market. On the latter, the government has just submitted a draft law on a new national payment system.

Foreign exchange markets. Important steps have recently been taken regarding the interbank foreign exchange market (MCI) with the elimination of exchange rate bands in foreign exchange auctions and the MCI, and measures to promote the practice of firm quotations among market participants. Further MCI reforms are needed with the adoption of an appropriate collateral system.

Mozambique’s financial sector reforms have also had positive spillovers to financial intermediation. The enhanced regulatory and supervisory framework and cleaned-up balance sheets—especially at BIM—have restored confidence in the banking sector, causing broad money growth to outpace nominal GDP growth and credit to the private sector to almost double in just two years (Table 4.2, Figure 4.8). These developments reflect structural factors associated with banks again taking on consumer credit (for example, credit cards and loans for the purchase of durable goods) and lending for industrial activities, as well as borrowing by domestic petroleum distributors related to the syndication of larger oil-import transactions. Lending to the agriculture and rural sectors remains minimal, however. Real lending rates and spreads have also been on a declining trend since 2002.23

Table 4.2.

Mozambique: Sectoral Contribution to Credit Growth,1 2005–06

(In percent of total credit growth)

article image
Source: Bank of Mozambique, Monthly Credit Survey.

Total and sectoral growth on a 12-month basis and derived from the Bank of Mozambique’s (BM’s) Monthly Credit Survey. This survey covers only a sample of the universe of financial institutions covered in the BM’s Monetary Survey.

Includes all loans to finance domestic and foreign trade activities such as oil and food imports.

Includes credit cards and consumer credit lines for vehicles and durable goods.

Nonclassified activities.

Figure 4.8.
Figure 4.8.

Mozambique: Bank Depth, 2000–06

Source: Bank of Mozambique.

Strengthening Mozambique’s Lending Environment

Framework for secured transactions. Additional measures include updating the legal framework for secured transactions. The recent enactment of a decree on urban land use and corresponding regulations was a first step in this direction; it is expected to facilitate the use of urban land as collateral by reducing the time and costs involved in transactions. Next steps in this area would be to tackle the modernization and linking of property registries and to implement a new Bankruptcy Law.

Enhancement of the scope and reliability of the credit registry. Reform is just starting in this area with assistance from the International Finance Corporation (the private sector arm of the World Bank Group) and the Millennium Challenge Corporation. With assistance from the IFC, the BM will conduct a feasibility study to examine options for increasing the scope and reliability of its credit registry.

MFIs have grown significantly in the past decade. During 2000–05 there was a fivefold increase in the number of clients and a sevenfold increase in outstanding loans (Figure 4.9). Consolidation also seems to be under way: 3 institutions out of about 35 control more than 68 percent of the active loan portfolio. However, MFI outreach is very limited. Rural areas account for less than 12 percent of total borrowers. Nonetheless, growth has been accompanied by a substantial improvement in performance. The percentage of MFIs with recovery rates considered acceptable (90 percent or above) increased from 16 percent in 1997 to about 30 percent in 2005. The access of the agriculture and rural sectors to banking and MFIs remains limited, however.24

Figure 4.9.
Figure 4.9.

Mozambique: Growth in Microfinance Activities

Source: Mozambique Microfinance Facility, 2006, Survey of Microfinance Operators.

Conclusions and Lessons for Sub-Saharan Africa

A prudent macroeconomic policy mix supported by monetary and financial sector reforms is critical to reducing inflation and ensuring financial stability in the early phases of stabilization. Macroeconomic stabilization in the context of very high inflation and a debt overhang, as in Mozambique in the early 1990s, requires fiscal consolidation to anchor inflationary expectations and avoid recourse to unsustainable domestic financing (and public debt) to maintain a credible nominal anchor. In this regard, relying only on sustained sterilized intervention could backfire, since such interventions limit the growth of the money supply but raise public debt. In sum, in countries with significant debt overhangs, purely money-based stabilization is unlikely to be effective without the support of fiscal consolidation. In addition, monetary and financial sector reforms, particularly reforms that address vulnerabilities in the banking system, as well as perseverance with prudent macroeconomic policies, can help maintain single-digit inflation rates and financial stability in the face of exogenous shocks, as demonstrated by Mozambique in the past few years.

A monetary targeting framework can be employed to maintain price stability and set the stage for adopting a more formal inflation targeting regime in the post-stabilization phase. Analysis of the monetary transmission mechanism in Mozambique suggests that base money could be a suitable operational instrument in controlling inflation in a country with shallow domestic financial markets, where interest rates may not yet have an identifiable transmission to prices, and there is a continuous need to mop up excess structural liquidity injected by a large fiscal deficit financed by buoyant aid inflows. Broader monetary aggregates could also provide a useful intermediate target but require further research, particularly regarding the stability of long-run money demand. Further research into the credit channel of monetary policy, interest rate determination, and degree of capital mobility in sub-Saharan Africa is also needed, given the difficulties of explaining real economic activity and exchange rate movements. As suggested in Chapter 5, “lite” inflation targeting that employs the instruments available to monetary policy to achieve a broad inflation objective is operationally close to the base money–targeting framework practiced in Mozambique and much of the rest of sub-Saharan Africa. It can be seen as an intermediate step toward switching to a full-fledged inflation targeting regime in which a specific level or range of expected inflation, instead of a monetary aggregate, is the intermediate target, and where the interest rate, rather than base money, is the operational instrument. In the interim, the focus should be on strengthening the monetary policy and liquidity management framework; providing greater central bank independence; and deepening debt, money, and foreign exchange markets—in other words, continuing to build the preconditions for a full-fledged inflation targeting regime.

Improving the outreach of the financial system remains a key challenge. With the soundness of the banking system restored, reforms to improve financial sector intermediation and access to financial services are assuming center stage. Policies and reforms in this area will need to continue strengthening the lending environment and facilitating the outreach of banks and MFIs and other private sector financial services to small and medium-size enterprises and rural areas. Looking ahead, it will be important to identify new quick wins. In this regard, special attention could be given to credit registries, particularly by expanding their scope and ensuring that their coverage increases in line with outreach growth. However, to ensure the soundness of an expanding financial system, outreach reforms will need to be accompanied by improvements in supervision practices, prudential regulations, and accounting standards beyond the banking sector. In the near term, introduction of risk-based supervision and international reporting standards in the banking sector will facilitate monitoring of capital adequacy during a period of strong credit growth. Rolling out International Financial Reporting Standards to the corporate sector will further reduce credit risk by improving the ability of the financial system to evaluate the quality of loan portfolios. Increasing attention will also need to be devoted to rural financing issues.

Mozambique’s relatively successful financial sector reform program has some important lessons to offer other low-income countries:

  • Getting the sequencing of reforms right is crucial. Mozambique’s reform experience highlighted the importance of making sure that first-generation reforms promoting financial sector liberalization and market-based banking systems are complemented at an early stage by efforts to strengthen the supervisory capacity and regulatory environment of the banking system to ensure the soundness of financial intermediation. Thus, reforms that ensure the soundness, efficiency, and stability of the banking system could be given higher priority than second-generation reforms aimed at promoting the access, outreach, and innovation of financial services.

  • Successful reform often requires making good use of brief windows of opportunity when there is broad political support for implementation. This widely recognized lesson is confirmed by Mozambique’s financial sector reforms as well as by business environment reforms (the latter are discussed in Chapter 9). Mozambique’s reform effort—particularly in the areas of banking supervision and reporting standards—was timely in the aftermath of bank restructuring in the early 2000s.

Bibliography

  • Adam, C., and S. O’Connell, 2005, “Monetary Policy in Sub-Saharan Africa” (unpublished; Washington: International Monetary Fund).

  • Agénor, P., and P. Montiel, 1999, Development Macroeconomics (Princeton, New Jersey: Princeton University Press, 2d ed.).

  • Alesina, A., and A. Drazen, 1991, “Why Are Stabilizations Delayed?” American Economic Review, Vol. 81 (December), pp. 117088.

  • Aron, J., and J. Muellbauer, 2001, “Interest Rate Effects on Output: Evidence from a GDP Forecasting Model for South Africa,” Staff Papers, International Monetary Fund, Vol. 49 (November), pp. 185213.

    • Search Google Scholar
    • Export Citation
  • Ball, L., and G. Mankiw, 1995, “Relative-Price Changes as Aggregate Supply Shocks,” Quarterly Journal of Economics, Vol. 110 (February), pp. 16193.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barnichon, Régis, and S.J. Peiris, 2007, “Sources of Inflation in Sub-Saharan Africa,” IMF Working Paper 07/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Barth, J.R., G. Caprio, and R. Levine, 2004, “Bank Regulation and Supervision: What Works Best?” Journal of Financial Intermediation, Vol. 13 (April), pp. 20548.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Batini, Nicoletta, and Anthony Yates, 2003, “Hybrid Inflation and Price-Level Targeting,” Journal of Money, Credit and Banking, Vol. 35 (June), pp. 283300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bernanke, Ben S., and Alan S. Blinder, 1992, “The Federal Funds Rate and the Channels of Monetary Transmission,” American Economic Review, Vol. 82 (September), pp. 90121.

    • Search Google Scholar
    • Export Citation
  • Bernanke, Ben S., and Ilian Mihov 1998, “Measuring Monetary Policy,” Quarterly Journal of Economics, Vol. 113 (August), pp. 869902.

  • Bernanke, Ben S., and others, 1999, Inflation Targeting: Lessons from the International Experience (Princeton, New Jersey: Princeton University Press).

    • Search Google Scholar
    • Export Citation
  • Calvo, G.A., 1983, “Staggered Prices in a Utility-Maximizing Framework,” Journal of Monetary Economics, Vol. 12 (September), pp. 38398.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Calvo, G.A., and C. Végh, 1999, “Inflation Stabilization and BOP Crises in Developing Countries,” in Handbook of Macroeconomics, Vol. 1C, ed. by J. Taylor and M. Woodford (Amsterdam: North-Holland), pp. 1531614.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Canzoneri, M., R. Cumby, and B. Diba, 2001, “Is the Price Level Determined by the Needs of Fiscal Solvency?” American Economic Review, Vol. 91 (December), pp. 122138.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Carare, Alina, Andrea Schaechter, Mark Stone, and Marc Zelmer, 2002, “Establishing Initial Conditions in Support of Inflation Targeting,” IMF Working Paper 02/102 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Castro, R., C. De Resende, and F. Ruge-Murcia, 2003, “The Backing of Government Debt and the Price Level,” Centre for Interuniversity Research in Quantitative Economics (CIREQ), Working Paper 16–2003 (Montreal: University of Montreal). Available via the Internet: http://www.cireq.umontreal.ca/publications/16-2003-cah.pdf

    • Search Google Scholar
    • Export Citation
  • Catão, L., and M. Terrones, 2005, “Fiscal Deficits and Inflation,” Journal of Monetary Economics, Vol. 52 (April), pp. 52954.

  • Chopra, A., 1985, “The Speed of Adjustment of the Inflation Rate in Developing Countries: A Study of Inertia,” Staff Papers, International Monetary Fund, Vol. 32 (December), pp. 693733.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Christiano, L., M. Eichenbaum, and C. Evans, 2000, “Monetary Policy Shocks: What Have We Learned and to What End?” in Handbook of Macroeconomics, Vol. 1A, ed. by J. Taylor and M. Woodford (Amsterdam: North-Holland).

    • Search Google Scholar
    • Export Citation
  • Christiano, L., M. Eichenbaum, and C. Evans, 2001, “Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy,” Working Paper No. 0107 (Cleveland: Federal Reserve Bank of Cleveland).

    • Search Google Scholar
    • Export Citation
  • Claessens, Stijn, and Luc Laeven, 2004, “What Drives Bank Competition? Some International Evidence,” Journal of Money, Credit and Banking, Vol. 36 (June, Part 2), pp. 56383.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Coe, D., and C. McDermott, 1997, “Does the Gap Model Work in Asia?” Staff Papers, International Monetary Fund, Vol. 44 (March), pp. 5980.

  • Cukierman, A., S. Edwards, and G. Tabellini, 1992, “Seigniorage and Political Instability,” American Economic Review, Vol. 82 (June), 53755.

    • Search Google Scholar
    • Export Citation
  • Engel, Charles, Mark C. Nelson, and Kenneth D. West, 2007, “Exchange Rate Models Are Not as Bad as You Think,” paper prepared for the NBER Annual Conference on Macroeconomics, Cambridge, Massachusetts, March 30–31.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, Stanley, 1993, “The Role of Macroeconomic Factors in Growth,” Journal of Monetary Economics, Vol. 32 No. 3, pp. 4566.

  • Fischer, Stanley, R. Sahay, and C. Végh, 2002, “Modern Hyper-and High Inflations,” Journal of Economic Literature, Vol. 40 (September), pp. 83780.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Frydl, Edward, 1999, “The Length and Cost of Banking Crises,” IMF Working Paper 99/30 (Washington: International Monetary Fund).

  • Galí, J., and M. Gertler, 1999, “Inflation Dynamics: A Structural Econometric Analysis,” Journal of Monetary Economics, Vol. 44 (October), pp. 195222.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ghosh, A., and Steven Phillips, 1998, “Warning: Inflation May Be Harmful to Your Growth,” Staff Papers, International Monetary Fund, Vol. 45 (December), pp. 672710.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hakura, Dalia S., 2005, “Are Emerging Market Countries Learning to Float?” IMF Working Paper 05/98 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hendry, D.F., 1995, Dynamic Econometrics (Oxford: Oxford University Press).

  • International Monetary Fund (IMF), 2003, World Economic Outlook, April 2003, World Economic and Financial Surveys (Washington).

  • International Monetary Fund (IMF), 2004, Republic of Mozambique: Ex Post Assessment of Mozambique’s PerformanceUnder Fund-Supported Programs, IMF Country Report No. 04/53 (Washington). Available via the Internet: http://www.imf.org/external/pubs/ft/scr/2004/cr0453.pdf

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2005, World Economic Outlook, September 2005, World Economic and Financial Surveys (Washington).

  • International Monetary Fund (IMF), 2007, Mozambique: Selected Issues, IMF Country Report No. 07/258 (Washington).

  • Kim, Soyoung, and Nouriel Roubini, 2000, “Exchange Rate Anomalies in the Industrial Countries: A Solution with a Structural VAR Approach,” Journal of Monetary Economics, Vol. 45 (June), pp. 56168.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kwon, G., L. McFarlane, and W. Robinson, 2006, “Public Debt, Money Supply, and Inflation: A Cross-Country Study and Its Application to Jamaica,” IMF Working Paper 06/121 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • La Porta, R., F. Lopez-de-Silanes, and A. Shleifer, 2002, “Government Ownership of Banks,” Journal of Finance, Vol. 57 (February), pp. 265301.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Leeper, Eric M., and Tack Yun, 2005, “Monetary-Fiscal Policy Interactions and the Price Level: Background and Beyond,” NBER Working Paper No. 11646 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Levine, Ross, 2005, “Finance and Growth: Theory and Evidence,” in Handbook of Economic Growth, Vol. 1A, ed. by P. Aghion and S. Durlauf (Amsterdam: North-Holland), pp. 865934.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liviatan, N., and S. Piterman, 1986, “Accelerating Inflation and Balance-of-Payments Crises, 1973–84,” in The Israeli Economy: Maturing Through Crisis, ed. by Yoram Ben-Porath (Cambridge, Massachusetts: Harvard University Press).

    • Search Google Scholar
    • Export Citation
  • Loungani, P., and P. Swagel, 2001, “Sources of Inflation in Developing Countries,” IMF Working Paper 01/198 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McKinnon, R.I., 1973, Money and Capital in Economic Development (Washington: Brookings Institution).

  • Mirakhor, Abbas, 1990, “Interest Rate Policies, Stabilization, and Bank Supervision in Developing Countries: Strategies for Financial Reforms,” IMF Working Paper 90/8 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Noy, I., 2005, “Banking Crises in East Asia: The Price Tag of Liberalization?” Asia-Pacific Issues Analysis No. 78 (Honolulu: East-West Center).

    • Search Google Scholar
    • Export Citation
  • Peersman, Gert, and Frank Smets, 2003, “The Monetary Transmission Mechanism in the Euro Area: More Evidence from VAR Analysis,” in Monetary Policy Transmission Mechanism in the Euro Area: A Study by the Eurosystem Monetary Transmission Network (New York: Cambridge University Press).

    • Search Google Scholar
    • Export Citation
  • Piñón-Farah, M., 1998, “Demand for Money in Mozambique—Was There a Structural Break?” IMF Working Paper 98/157 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Posen, A., 1995, “Central Bank Independence and Disinflationary Credibility: A Missing Link?” Staff Report No. 1 (New York: Federal Reserve Bank).

    • Search Google Scholar
    • Export Citation
  • Sargent, T.J., and N. Wallace, 1981, “Some Unpleasant Monetarist Arithmetic,” Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 5, No. 3, pp. 117.

    • Search Google Scholar
    • Export Citation
  • Saxegaard, Magnus, 2006, “Excess Liquidity and Effectiveness of Monetary Policy: Evidence from Sub-Saharan Africa,” IMF Working Paper 06/115 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shaw, E.S., 1973, Financial Deepening in Economic Development (New York: Oxford University Press).

  • Shleifer, A., and R. Vishny, 2002, The Grabbing Hand: Government Pathologies and Their Cures (Cambridge, Massachusetts: Harvard University Press).

    • Search Google Scholar
    • Export Citation
  • Sims, Christopher, 1980, “Macroeconomics and Reality,” Econometrica, Vol. 48 (January), pp. 148.

  • Sims, Christopher, 1986, “Are Forecasting Models Usable for Policy Analysis?” Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 10, No. 1, pp. 116.

    • Search Google Scholar
    • Export Citation
  • Sims, Christopher, and Tao Zha, 2006, “Does Monetary Policy Generate Recessions?” Macroeconomic Dynamics, Vol. 10, No. 2, pp. 23172.

  • Stone, Mark, 2003, “Inflation Targeting Lite,” IMF Working Paper 03/12 (Washington: International Monetary Fund).

  • Woodford, M., 2001, “Fiscal Requirements for Price Stability,” Journal of Money, Credit and Banking, Vol. 33, No. 3, pp. 669728.

  • World Bank, 2003, “Mozambique Industrial Performance and Investment Climate 2003” (Washington).

1

Alternatively, the central bank could tie its currency to another country’s currency to maintain price stability, but that option is not under consideration for Mozambique, which aims to maintain a flexible exchange rate regime (to absorb shocks and maintain a comfortable level of international reserves), and is, therefore, not discussed here.

2

A similar line of reasoning lies behind the fiscal theory of the price level (Canzoneri, Cumby, and Diba, 2001; and Woodford 2001). This theory posits that increased government debt adds to household wealth and, hence, to demand for goods and services, leading to price pressures.

3

However, they do not detect a strong positive connection between deficits and inflation in low-inflation advanced economies, which may be explained in part by the greater autonomy and credibility of these economies’ monetary policy, their deeper financial markets, and other institutional constraints that link public borrowing more closely to tax and spending smoothing.

4

The relationship between deficits and inflation was robust to the inclusion of other explanatory variables, such as oil prices and degree of trade openness, as in Catão and Terrones (2005).

5

In contrast, in 13 major advanced economies, none of the explanatory variables, except lagged inflation, show significant short-term associations with inflation.

6

The exchange rates are also included in the robustness test to control for possible biases from exchange rate volatility on the debt dynamics. The directions of the impulse responses remain unchanged in an alternative VAR that includes the exchange rate as an endogenous variable.

7

See the section “Monetary and Financial Sector Reforms.”

8

Across sub-Saharan Africa—but particularly outside the CFA zone—there has been a steady decline in domestic deficit financing. The overall decline was similar for poststabilization economies, with central bank credit showing the biggest decline.

9

According to Adam and O’Connell (2005), the contraction in domestic credit since the 1990s has been associated with a steady decline in inflation in sub-Saharan Africa. Outside the CFA zone, inflation has been consistently in the single digits, although the averages mask a number of cases where fiscal dominance persisted and inflation remained out of control—for example, Angola, the Democratic Republic of the Congo (until 2002), and, more recently, Zimbabwe—and a number of countries where inflation remained at persistently high levels, at least by contemporary standards (Nigeria, The Gambia, Ghana, Malawi, and Zambia).

10

It is now widely accepted that the primary role of monetary policy is to maintain price stability (IMF, 2005; and Batini and Yates, 2003), which is defined by many as an annual rate of inflation in the low single digits in industrial countries (Bernanke and others, 1999) and in the single digits in low-income countries (Fischer 1993; and Ghosh and Phillips, 1998).

11

There is also the question of how the instruments available for conducting monetary policy should be deployed in shock-prone post-stabilization economies in sub-Saharan Africa such as Mozambique, while these economies maintain a commitment to low and stable inflation (and reduced macroeconomic volatility, more generally). This question is discussed in Chapter 5.

12

Quarterly GDP series were only recently made available in Mozambique. However, the series is still not short and subject to revisions; therefore, a quarterly GDP proxy was constructed statistically by estimating the correlates of real annual GDP and using predictions based on the quarterly explanatory variables.

13

An analysis of the time-series properties of the variables revealed that the variables are integrated of order one or I(1). Therefore, the series are first-differenced for estimation purposes to avoid the spurious and inconsistent regression problem (Hendry, 1995).

14

Instead, see Piñón-Farah (1998) for analysis of the long-run demand for money in Mozambique.

15

This may be explained by the greater share of nontradable goods in core inflation.

16

It should be noted, however, that Chapter 6 reports that foreign exchange intervention operations have a significant impact on the nominal exchange rate, indicating that the central bank’s actions could have an impact on the exchange rate, even though their impact on base money is not discernible, possibly because of sterilization operations.

17

Stone (2003) defines inflation targeting “lite” regimes as ones where the central bank “announce[s] a broad inflation objective but owing to [its] relative low credibility [it is] not able to maintain inflation as the foremost policy objective. Their relatively low credibility reflects their vulnerability to large economic shocks and shallow financial markets and a weak institutional framework.”

18

See IMF (2004) for a more detailed account of reforms undertaken in the context of IMF-supported programs during this period.

19

The FSAP is a joint IMFWorld Bank initiative introduced in 1999. FSAPs are undertaken at the request of a member country for the purpose of identifying the strengths and vulnerabilities of the country’s financial system, determining how key sources of risk are managed, ascertaining the financial sector’s development and technical assistance needs, and helping the country prioritize its policy responses.

20

See IMF (2007) for details on the Mozambican authorities’ forward-looking financial sector strategy.

21

NBFIs in Mozambique were deemed small, nontransparent, and subject to limited institutional and regulatory oversight. As such, they could mask sizable contingent fiscal liabilities. The situation was particularly critical in the insurance and pension sectors, where supervisory capacity was diluted and, in the case of pensions, weak. In particular, supervisory capacity in the insurance sector was shared by the Ministry of Finance (insurance) and the National Social Security Institute (INSS). Supervision of pensions was weakened by the fact that the Ministry of Finance had both an administrative and a prudential mandate.

22

The strong decline in credit denominated in foreign currency was driven, in large part, by a prudential measure introduced in July 2005 that required loan loss provisioning to cover 50 percent of foreign currencydenominated loans to nonexporters. This decrease was, however, accompanied by an increase in applications for private external financing.

23

Issues related to financial depth are discussed further in Chapter 9 in the context of reforms in the business environment designed to improve the access of small and mediumsize enterprises and rural areas to financing.

24

About 62 percent of MFIs and 50 percent of micro borrowers are concentrated in the Maputo-Matola urban corridor; rural areas account for less than 12 percent of borrowers and only 5 percent of the total active portfolio. To some extent, this disparity in microfinance provision seems to reflect the large concentration of nonagricultural informal activities in the province of Maputo or Maputo City. Preliminary data for 2005 on informal sector activities published by Mozambique’s national statistical agency show that of the nonagricultural informal sector activities, 42 percent of those involved in trading and 48 percent of those involved in services (excluding transport) are located in those areas.