Implementing Inflation Targeting in Sri Lanka: The Fiscal Challenge

The traditional Quantity Theory of Money argues that inflation is always and everywhere a monetary phenomenon. However, the recent Fiscal Theory of Price Level presents a formidable challenge to this view by recognizing that fiscal influence plays a key role in determining inflation. Some authors contend that fiscal influence over monetary policy could even break down Inflation Targeting (IT) regimes. This study pre-empts the challenge of fiscal dominance in implementing an IT framework in Sri Lanka. It argues that the balance of fiscal-monetary relationship is tilted towards fiscal dominance despite the protections – although limited in nature included in the existing legal framework. Moreover, characteristics of the fiscal-monetary relationship go well beyond safe terrains vis-à-vis international best practices. It is important that the central bank and the government agree on phasing out fiscal influence in ‘spirit’ in addition to enshrining legal safeguards in a piece of legislation. The period of phasing out should form an integral part of the action plan to implement IT. Failure to do so can have implications on the success of implementing IT and adversely impact the credibility of the central bank.


Background
Friedman's assertion that 'inflation always and everywhere a monetary phenomenon' remains influential in Monetary Economics even today. While there have been studies lending strong support, this view has not gone unchallenged. 'Some Unpleasant Monetary Arithmetic' by Sargent and Wallace (1981) and the 'Fiscal Theory of Price Level' promoted by Leeper (1991), Sims (1994 and Woodford (1995) and the subsequent writings have emphasized the significance of fiscal policy and debt level in to the debate on determination of price level.
For practitioners who work on Inflation Targeting (IT), these theoretical expositions provide food for thought. Ever since New Zealand pioneered IT, such frameworks showed a tendency to focus more on legal and institutional frameworks pertaining to the central banks for the implementation of IT. Perhaps, this may be due to fair reasons as it is very important to get 'that' house in order, which is responsible for delivery of the inflation target. However, relatively mixed outcomes have led some authors to increasingly focus on other related factors adversely influencing IT implementation such as fiscal dominance and weak financial systems.
While the literature suggests that a strong fiscal and debt position could be helpful, it is not an essential condition for the implementation of IT frameworks. However, monetary policy independence or the absence of fiscal dominance over monetary policy appears to be an important ingredient for the successful implementation of IT. A strong fiscal position that rules out the need to monetize the deficit would reinforce; in general, a country needs to strengthen its fiscal position prior to the transition to inflation targeting.
in which the Ministry of Finance engaged in monetary policy related matters in public. The fiscal budget 2012 forced a three percent depreciation overnight with a view to 'encourage exports'. In the respective budget speech, the Finance Minister remarked that "I propose to depreciate our exchange rate by 3 percent with immediate effect, to correct this disadvantageous position and to encourage our exports. I expect that the Central Bank will adopt appropriate modifications to the Monetary Policy accordingly". Moreover, it appears that key decisions on capital of the CBSL is subject to the Ministry of Finance's discretion; Finance Minister at budget 2014 said that "The performance of the Central Bank of Sri Lanka ……. in recent years, has been exemplary.
Considering Central Bank's responsibilities and the recent trends in the economy, I propose to increase its capital to Rs.50 billion". For another, budget speech 2015 purported to make an exception to the statutory reserve ratio of the CBSL: "Central Bank will exempt the statutory reserve requirement on individual deposits maintained by the elderly people in all commercial banks". Although it was not implemented, budget 2016 required the CBSL to provide a 100 guarantee to depositors of the Finance Companies: "…. in order to provide the depositors with a sense of comfort and security, the Central Bank of Sri Lanka will give a 100 percent guarantee on all deposits of all the registered finance companies by end January 2016". Similarly, fiscal budgets openly solicit revenues from the CBSL through the text of the budget speech. 1 As an interim arrangement, the CBSL has adopted an enhanced monetary policy framework, which, it claims, has features of both Traditional Monetary Targeting (TMT) and Flexible Inflation Targeting (FIT) frameworks. Under this, the CBSL is said to focus on stabilizing inflation in mid-single digits over the medium term, while supporting growth objectives and flexibility in exchange rate management. This move has been backed by the International Monetary Fund (IMF) facilitated 2016 program.
Hence, the present study is an attempt to understand the fiscal-monetary relationship in Sri Lanka within the existing legal and institutional framework, and its potential consequences on the proposed move to an IT framework. The findings of the study will be of practical significance for the CBSL and the government in general in crafting the monetary-fiscal relationship and strengthening the institutional framework that would facilitate a smooth transition. It is important that CBSL commits to an IT frameworkin which it would be held accountable for achieving the targetswith adequate research on challenges that it will be faced with.
Section 2 of this article reviews the literature on IT and existing theoretical expositions on the impact of fiscal policy and its dominance on IT implementation. The Section 3 looks at fiscal and debt dynamics and explores the fiscal and monetary relationship in the Sri Lankan context, given their importance for IT implementation.

Introduction to Inflation Targeting
The evolution of IT, at least in part, was a response to the search for a monetary management framework after Bretton Woods collapsed. When targeting of a fixed exchange rate no more became an option for monetary management, the central banks world over started looking for a potent alternative. Although monetary targeting became popular among some countries, the success of continued implementation of such frameworks fell below expectations (Rose 2007).
Moreover, inability to define clear performance measures and systems of accountability was a key weakness of many of the varying versions of monetary management frameworks. IT was meant to answer many of these problems while providing clear direction to contemporary monetary management practice. Bernanke et al. (1999) define IT as "…a framework for monetary policy characterized by the public announcement of official quantitative targets (or target rages) for the inflation rate over one more time horizons, and by explicit acknowledgement that low, stable inflation is monetary policy's primary long-run goal". IT was first introduced by New Zealand in its efforts to implement governance and economic related reforms across the government departments. The Reserve Bank of New Zealand in 1989 established the overall policy framework for IT that included (a) price stability as the primary function of the central bank, 2 which was clarified by inflation targets; (b) central bank independence; and (c) accountability through making the target public and holding the governor responsible for achieving it. Ever since, IT has continued its expansion among both advanced as well as Low and Lower Middle-Income Countries with over 35 countries formally adopting IT by March 2015 (IMF;. A survey carried out by Laurens et al. (2009) suggests that 95 percent of 181 countries covered have explicitly identified price stability as one of the primary objectives of monetary policy and that price stability receives priority when there are more than one objectives, an ingredient required in the transition towards IT. Mishkin (2004) establishes five components as forming IT framework: 1. the public announcement of medium-term numerical targets for inflation; 2. an institutional commitment to price stability as the primary goal of monetary policy, to which other goals are subordinated; 3. an information inclusive strategy in which many variables, and not just monetary aggregates or the exchange rate, are used for deciding the setting of policy instruments; 4. increased transparency of the monetary policy strategy through communication with the public and the markets about the plans, objectives, and decisions of the monetary authorities; and 5. increased accountability of the central bank for attaining its inflation objectives.
In summary, an IT regime could be identified as one which formally announces an inflation target and adjusts a policy instrument in a way consistent with the target. In such a system, performance of a central bank could be measured against the target and make it accountable for achieving the target.
Many economists attach IT's significance to its transparency (Sims 2005;Walsh 2009). Arguing the case for not implementing 'spurious technocratic' solutions for 'political and institutional pathologies' that lead to inflation (or deflation), Sims (2005) infers "inflation targeting is in most countries an improvement in the monetary policy regime. But the improvement comes from its being a step toward goal and model transparency. The central bank should probably everywhere be charged with making projections of inflation, laying out policy actions that could stabilize inflation, and either taking those actions or explaining why it cannot and who could".
'Is IT necessary for good monetary policy? ' Walsh (2009) answers this question in negative; in principle, other regimes could also provide the required nominal anchor while ensuring the flexibility needed to promote overall economic stability. However, he argues that "a system of maintaining a medium-run focus on controlling inflation, communicating clearly with the public about the ultimate objectives of monetary policy, and providing a measure of accountability inflation targeting dominates the alternative choices". The experience with IT for the last 25 years highlights that decline in inflation since 1990's among industrialized countries coincided with the adoption of IT, making it difficult to separate out direct benefits of an IT framework. However, IT seems to be beneficial for developing countries with significant reduction of inflation and inflation volatility. The fact that no country has left the club except for Finland and Spain (who abandoned IT to join Eurozone) once adopting an IT framework speaks for its success (Walsh 2014;Rose 2014). Moreover, IT has been able to reduce inflation volatility compared to non-IT frameworks (Gürkaynak, Levin and Swanson 2010). Moreover, Rogers (2009) finds that, although inflation target ranges are missed frequently in most of the inflation targeting countries, the inflation and growth performance under inflation targeting compares very favorably with performance under alternative frameworks.
IT is not devoid of criticism. Critics argue that overly focus on inflation or pursuing it as the only objective of monetary policy would inevitably mean that other objectives will be neglected and because of that some authors call for abandonment of reforming IT (Friedman 2004). This growing concern along with the impact of the Global Financial Crisis (GFC) that amplified the need to pursue other objectives has led to reforms and a resultant notion of FIT. In a FIT the central bank aim at keeping inflation near the target in the medium-term while taking into account the impact on real economy, interest rate, or exchange rate volatility, and financial system stability in setting the pace of adjustment in policy implementation (Woodford, 2013;Svensson, 2009). In the strictest sense, it appears that no central bank is a strict targeter. All inflation targeters behave in ways consistent with a concern for both inflation and real economic stability. It appears that IT central banks conduct flexible inflation targeting rather than strict inflation targeting. Laurens et al. (2015) suggest that implementation of an IT framework should be undertaken at the pace deemed appropriate for each country context. The time consumed in the process significantly vary across jurisdictions. The histogram drawn using data collected by Rose (2007) for a sample of 25 countries indicates an average time taken of 7.8 years to complete implementation of a IT framework in full.

2.1
Important building blocks in an IT framework Carare et al. (2002) identify the initial conditions in support of an IT framework into four groups.
While admitting that these conditions are not strict pre-requisites, they emphasize that these may be necessary in the short to medium-term when the framework is implemented. 3. The financial system is developed and stable enough to implement the framework (f) There should be sufficient financial stability to enable monetary policy to pursue the inflation targets and not be sidetracked by concerns about the health of the financial sector.
(g) Financial markets should be sufficiently well developed to enable monetary policy to be implemented using market based instruments, and to ensure that the conduct of monetary policy is not complicated by weaknesses in financial market infrastructure.

The need for proper tools to implement monetary policy in support of the inflation target (h) A central bank should be in a position to influence the inflation through its policy
instruments and have a reasonable understanding of the links between the stance of policy and inflation.
(i) Exchange rate objectives must be clearly subordinated to the inflation target. Therefore, the central bank should endeavor to make clear that foreign exchange market interventions and changes in the policy interest rate intended to influence the exchange rate are only aimed at smoothing the effects of temporary shocks.

Fiscal position and IT: Theoretical and empirical dimensions
Debate of the interaction between monetary and fiscal policies in the determination of inflation dates to pre-IT era. Backed by the Quantity Theory of Money, Friedman and Schwartz (1963) argue that 'inflation is always and everywhere a monetary phenomenon'. While describing what monetary policy can do and can not do, Friedman (1968) asserts that a central bank could control inflation rate, especially in the long run.
Challenging Friedman, Sargent and Wallace (1981) argue that Friedman's list of the things that monetary policy cannot permanently control may have to be expanded to include inflation. They show how the capacity of one policymaker (fiscal) to commit to a given policy path ultimately forces the other (monetary) pursue a policy course needed to meet the solvency constraint of the government. According to Sargent and Wallace, the monetary authority must generate seigniorage revenues to secure government solvency when the fiscal operations stick to a given path. A tight monetary policy to keep inflation down cannot last and must ultimately give in to higher inflation in the longer run, in the 'unpleasant monetarist arithmetic' of Sargent and Wallace (1981). Their theoretical exposition lends to the phenomena that a weak fiscal position could harm the ability of the monetary policy to pursue stabilization of inflation.
Contributions on Fiscal Theory of Price Level presented by Leeper (1991), Woodford (1995) sees the inflation as the equilibrium price level that makes the real value of nominally denominated government liabilities equivalent to the present value of expected future government budget surpluses. He argues that an increase in the price level reduces the real value of the private sector or net government liabilities of the government leading to a wealth effect that reduces private sector demand for goods and services. While Woodford does not completely deny the importance of Quantity Theory of Money in the determination of inflation, his theory underscores the significance of stability of expectations regarding future government policy for stability of the price level. The effects of "fiscal policy changes on the equilibrium price level are largely independent of any changes in the path of the money supply that might be associated with them".
These challenges to the traditional Quantity Theory -which underscore the significance of considering fiscal and debt position when committing to an IT regime -are not without merit.
Some authors argue that IT cannot work when a country is having high debt. Analyzing Brazilian experience of inflation targeting, Blanchard (2004) concludes that "in an environment of high debt, inflation targeting may not work well, or indeed may not work at all". A central bank engineered real interest rate increase aimed at the target inflation rate may lead to higher debt and higher risk of default, a depreciation, and, so, to higher inflation rather than lower inflation. He emphasizes that in such an environment, "the way to fight inflation is through fiscal policy, not through monetary policy".
While a strong fiscal position is not an essential pre-condition for implementation of IT frameworks, it is very important that monetary policy is not dominated by fiscal operations (Debelle, 2001;Masson et al., 1997;Mishkin 2001). Kumhof et al. (2008) argue that fiscal reform in developing countries is an 'indispensable step' before implementing inflation targeting regimes.
A strong fiscal position that rules out the need to monetize the deficit reinforces the implementation of IT. In general, a country needs to strengthen its fiscal position prior to the transition to inflation targeting. (Carare et al. 2002). 3 Hammond et al. (2009) argue that in many of the developing countries, long term fiscal discipline is weak and 'monetary policy is often an 'adjunct to fiscal policy'. In some countries, this takes the form of the central bank being statutorily under the purview of the finance ministry. In some countries where the central bank is in principle independent, there is still the reality that it can be influenced by various political forces, especially the finance ministry. Similar conclusions have been reported by Resende (2007) for IMF member countries. An unsustainable fiscal policy, characterized by high fiscal deficit, public debt, and debt service along with high contingent liabilities could severely constraint the successful implementation of an IT framework. Loyo (2000) shows that monetary policy is ineffective in containing inflation when the fiscal policy backing is absent. In order to control inflation, government finances must be sustainable (Leeper, 1991;Sims, 1994;and Woodford, 1994and Woodford, , 1995and Woodford, and 2001 The Turkish experience of the transition to IT demonstrate the significance of choosing the right time although it is counterproductive to be 'perfectionist'. Turkey took time while receiving support from IMF. In 2002, the central bank of Turkey announced that it will move towards 'implicit IT' and that it will "openly initiate the inflation targeting regime whenever the conditions emerge. We had to postpone the introduction of inflation targeting due to concerns about the sustainability of domestic debt….". It is noted that  (Ersel et al.: 2008;Eroglu et al.: 2017).
The vast majority of central banks who have adopted IT seem to have a conducive fiscal environment during the time of transition. Out of a sample of 29 IT countries, only seven had an average of above five percent fiscal deficits for the three years immediately prior to transition to IT while only six countries had a public debt to GDP ratio of more than 60 percent (Figure 1).
Most of the IT countries, have not been experiencing fiscal stress at the point of transition.

Sri Lanka's fiscal landscape and its potential impact on IT
According to the literature, a weak fiscal position, high debt and budget rigidity is an enticement for fiscal operations to look for support from the central bank. While a strong and credible fiscal rules framework could act as a deterrent to a fiscal policy with limited discipline, weak institutional and instrumental independence makes the central bank vulnerable for fiscal dominance.
In this context, the present study evaluates the (a) past fiscal-monetary relationship; (b) fiscal landscape in relation to fiscal and debt dynamics; (c) compliance with fiscal rules framework and attempts to pre-empt the influence that the circumstances will have on the implementation of an IT framework in Sri Lanka.

Fiscal-monetary relationship in Sri Lanka
In  In application, CBSL credit to the government seemed to have exceeded the 10 percent limit for sustained periods in the past (Figure 3). One inference coming out of this analysis is that the CBSL may have helped the government at difficult times (e.g. 2001peak of terrorist activity, 2008/2009 peak of war, 2011/12energy subsidies, balance of payment difficulties). Interestingly, such assistance was followed by IMF programs/disbursements in response to external sector difficulties and loss of reserves created at least in part by rapid monetary expansion. 6 Continued fiscal financing even going over the statutory limits, arguably, is an indication that the CBSL has been subject to fiscal dominance. Figure 3: CBSL credit to government as a share of budgeted revenue vs statutory limit (including both provisional advances and bills purchased in domestic operations) Source: Central Bank of Sri Lanka and author's calculations Note: CBSL credit to government in the graph includes both provisional advances and lending through treasury bills based on the argument that MLA imposed limit on provisional advances, in spirit, is a restriction on CBSL financing of government. Lending through treasury bills was not envisaged as an instrument either by the MLA or the Exter Report.
times; and it is cited that these securities were not used due to lack of liquidity. It could be argued, however, that liquidity on CBSL securities would pick up overtime when the market participants get used to them. These securities could easily be seen as secure instruments in par with government securities and play the liquidity management role while minimizing the influence from the fiscal front. 6 It may be difficult to conclude that periods of sustained monetary support to fiscal operations have led to Balance of Payment difficulties without a comprehensive study of the effects of increased money supply and resultant pressures on the exchange rate requiring CBSL intervention in the forex market and other global and local developments. Given the scope of the study, this aspect is not covered here.  The graphical representation affirms this position. The Figure 4 plots the net injection (positive) or absorption (negative) against the Treasury bill stock held by the CBSL. When the CBSL injects liquidity, the Treasury bill stock is expected to reflect the same movement, at least until February 2014 after which Standing Deposit Facility was uncollateralized by the CBSL given its implicit zero risk on local currency obligations. Although this expected almost one-to-one relationship is not visible, the Treasury bill stock has indicated a clear link with the CBSL credit to government ( Figure 5). Moreover, issuance of sovereign bonds also seems to have decelerating effect on CBSL credit to government. Although not conclusive, it appears that issuances of sovereign bonds have helped to manage the fiscal pressures for a few months and led to a reduced level of government securities held by the CBSL during such period.   The above analysis of OMOs leads to an inference that the Treasury bill stock held by the CBSL has limited facilitation effect on market liquidity while it has mainly been used for providing fiscal support in a low revenue context. This argument is further strengthened by the fact that the CBSL did not conduct daily auctions for extended periods when the market was experiencing excess liquidity. 7 Although such an action could be used as a first-line defence against deterioration of the central bank balance sheet due to increased interest payments on absorptions (Adler et al.

2016), it could reduce effectiveness of monetary policy.
This fiscal monetary relationship appears to be the result of a challenging fiscal and debt landscape, a weak fiscal rules regime in a context of the institutional weaknesses. The Sections 3.2 and 3.3 attempt to briefly describe the sources of fiscal influence.

Sri Lanka's fiscal position
Sri Lanka's fiscal operations are faced with many challenges. Persistently high overall fiscal deficits 8 , debt levels, interest cost and future gross financing requirements are key characteristics of fiscal operations ( Figure 6). The budget is relatively small as a share of GDP by international standards 9 given the income level of the country. This is mainly due to the government's efforts to maintain the deficit at acceptable levels in the wake of low and declining revenues as a share of GDP experienced in the last two decades 10 (Figure 7). The expenditure-focused consolidation efforts have led to a lean and rigid budget with non-discretionary expenditures in salaries, wages, pensions and interest payments account for 60 percent of the total (Figure 8). Limited fiscal space, 11 which is linked to the budget rigidity, has reduced allocations for important economic sectors including health education and social protection over time.
7 High liquidity was a common experience for emerging market central banks that led to deterioration of the balance sheets between 2009 and 2014. Central banks in the emerging markets had to absorb increased liquidity by paying substantial interest volumes. 8 On average 6.2 percent of GDP between 2015-2017 9 The total government expenditure is less than 20 percent of GDP since 2010, down from around 30 percent of GDP in early 1990's 10 Tax revenue to GDP that was 24.2 percent of GDP in 1978 declined to 14.5 percent in 2000 and bottomed at 10.1 percent in 2014. A long list of exemptions on VAT; generous tax incentives, concessionary rate and other exemptions on income tax led to erosion of tax revenues. Informality and weaknesses in tax administration also contributed to reduction of tax collection. 11 Fiscal space could be identified as "room" in a government's budget that allows it to provide resources for a desired purpose without jeopardizing the sustainability of its financial position or the stability of the economy

Sri Lanka's fiscal rules framework
The Fiscal Management Responsibility Act of 2003 (FMRA) and its subsequent amendments specify Sri Lanka's fiscal rules regime with limits defined on public debt to GDP ratio, fiscal deficit to GDP ratio and guarantees to GDP ratio on a rolling basis ( Table 2). The framework identifies responsible fiscal management, prudence in debt management and public scrutiny over fiscal affairs as key priorities to be achieved through implementing this framework.  Sri Lanka's fiscal rules framework appears to have operated with limited credibility. The problems associated with discretionary policy have not allowed achieving goals set out in the fiscal rules framework in Sri Lanka. Since enacted in 2003, the country has not been able to adhere to the fiscal rule on budget deficit despite continued fiscal consolidation efforts by successive governments (Figure 12). The amendment in 2013 postponed the compliance with the public debt related fiscal rule, in anticipation of a breach in the particular year ( Figure 11) while the fiscal rule related to the guarantee limit was changed in both 2013 and 2016 when it was about to touch the maximum limit ( Figure 13). These changes adversely impact credibility of the framework, and continued non-adherence poses challenges to implementation of an IT framework.

Conclusion
A monetary authority has limited space to acknowledge even if it is adjunct to fiscal operations.
However, implementing an IT regime could expose a monetary authority unlike a TMT by making it accountable for achieving an inflation target while fiscal operations continue to complicate monetary policy operations. When assessed using international parameters, Sri Lanka's balance of deficit/debt and legal/institutional dynamics is tilted towards fiscal dominance. Providing provisional advances based on the revenue budgeted for the forthcoming year, not charging any interest on such advances, purchase of treasury bills in the primary market are characteristics that could go well beyond international practices, even in the context of the developing world. In addition to these systemic issues, there have been sporadic interventions by the fiscal budget as well; for example, overnight depreciation of the currency by three percent in 2012.
Given that the success of an IT framework is vulnerable to the impact of fiscal operations, it is important that the CBSL and the government agree on phasing out fiscal influence in 'spirit' in addition to enshrining legal safeguards in a piece of legislation before implementing IT. 12 The period of phasing out should form an integral part of the IT action plan. Failure to do so can have implications on the success of implementing IT and adversely impact the credibility of the central bank.
12 Cabinet approval for relevant Amendments to the MLA by 2018 for submission to Parliament by March 2019 is a structural benchmark of the IMF program.