By Sameer Sharma
The most successful central bankers will tell you that the key to the success of monetary policy is that the monetary policy framework needs to be well understood, quantifiable and the market also has to believe that the central bank will do whatever it takes in order to achieve its pre-set objective. When it comes to the market itself, the infrastructure to transmit monetary policy to prices and output must also be developed enough, else it will not work well.
More than 53 years after independence, the Bank of Mauritius is still trying to come up with a credible and quantifiable framework which will deliver. The Bank of Mauritius officially has a mandate to maintain price stability and orderly economic development. What many do not ask however is what does this actually mean? Is this a 3% or 4% or 5% inflation rate? What about output and the exchange rate? The quarterly interventions by the many local economists in the media about where interest rates should be all sound confusing to this author given that the level of interest rates would depend on the quantification of what is meant by price stability (explicit inflation target for example) and by what is meant by orderly economic development.
The reality is that such objectives are so vague that one could essentially argue to set interest rates or any other monetary policy tool metric to almost anything. Without a quantifiable framework, such discussions are interesting but are not very useful. The objective of this article is to not only explain why the central bank is still soul searching but more importantly how a new monetary policy framework could work taking into account the post-2020 Bank of Mauritius balance sheet constraints, the current state of development of the secondary bond market and the planned roll-out of a central bank backed digital currency.
Structural excess liquidity
While ministers and prime ministers cannot ignore the electoral cycle, central bankers are meant to be equipped to operate independently beyond such considerations. Pressure from politicians to align monetary policy to their overall fiscal objectives will always come, but the key is for the central bank to be able to resist such pressures. In order to ensure that a central bank can focus on the long term picture and on its medium term target, the contract terms of the governor along with his/her two deputies should be much longer than the typical 3 year renewal term which falls well within the electoral cycle. Beyond the pressure that will invariably come from politicians, by the time a governor is able to understand what should be done and how to do it, his/her mandate may just come to an end, and it may be too late.
It is also important to ensure that monetary policy committees are manned by economists with at least an advanced degree in monetary economics. We should not be shy about seeking some of this competence from abroad. Monetary policy setting is not like how it was done back in the Soviet days where so many barristers are needed. Monetary policy has little to do with accounting and law.
It remains important for the government via a finance minister in a democracy however to help set the medium term monetary policy objective with the central bank. The Bank of England and Her Majesty’s Treasury, for example, agree on a clear medium term inflation target, and the central bank is then accountable for meeting its objective. This is not the case in Mauritius.
The worst enemy of a central banker is when he/she starts to think that multiple objectives can be met. There is actually no trade-off between growth and inflation over time. Price stability is good for economic growth. Monetary policy in the best of cases aims to smooth the business cycle and looks at medium term growth forecasts versus potential growth, all in the context of price stability.
The central bank also has to be clear about which channel it will use in order to achieve a clear and quantifiable medium term objective. In a nutshell, if a central bank’s primary tool to achieve a medium term inflation target is through the interest rate channel, then it should not have multiple conflicting objectives. For example, if a central bank intervenes regularly in the foreign exchange market to buy foreign currency, it adds liquidity into the system, which may counter a tighter stance it may have taken on rates in a context of higher inflation expectations. Sterilized interventions can in themselves be quite costly. Trying to do too many things means achieving little in the end.
Central banks cannot technically become insolvent as long as their liabilities are in local currency, but a central bank, especially an emerging market one, let alone a frontier market central bank, should avoid a situation where it has negative equity and has to print money or depend on favourable exchange rate movements in order to either meet its liabilities or appear to have a stronger balance sheet than it really has. One big challenge frontier market central banks like the Bank of Mauritius face is structural excess liquidity which often is the result of years of accumulated unsterilized foreign exchange reserves accumulation.
Central banks need strong balance sheets in order to be perceived as being credible.
This is because issuing central bank paper to mop up excess liquidity can be expensive while cost sharing agreements between the government and the central bank may be hard to implement in practice given that governments would prefer to spend the money elsewhere. This lack of coordination complicated by multiple competing objectives, which creates structural excess liquidity, essentially pushes such central banks to widen their repo corridors or to allow market rates to hover outside an already wide corridor. This leads to a weak transmission mechanism and near zero credibility for the central bank.
The Bank of Mauritius may have more than MUR 126 billion in Bank of Mauritius instruments, but excess liquidity remains well above MUR 26 billion. Besides, the current outstanding amount, when compared to the January 2020 level, is not very impressive. Increasing interest rates while maintaining market rates within the corridor would be an expensive task.
Central banks need strong balance sheets in order to be perceived as being credible. Economic capital should be appropriately sized, based on the risks a central bank carries on its balance sheet. Economic capital should be a function of market risks inclusive of foreign exchange rate risk, operational risk, contingency buffers, monetary policy considerations, credit risk, the liquidity risk from investing in non-liquid assets and even reputation risk. Such considerations of course include having enough capital to cover for the high level of credit risk from the investments of the Mauritius Investment Corporation (MIC) into distressed assets, something central banks do not include on their balance sheets like the Bank of Mauritius has. This is all complex stuff, which is why only the best tend to work at central banks.
On the foreign exchange reserve side, global bond yields are at historical lows. The central bank must be able to generate enough revenues to more than credibly offset sterilization costs. The bulk of its revenues as at date come from foreign exchange reserves and the MIC.
The Mauritian local bond market has poor turnover given its high degree of bond market fragmentation. This is partly explained by a lack of a modern debt management strategy, which has led to too many small government bond issues.
For monetary policy to effectively transmit from the policy rate to prices and output over time, the market ecosystem which is essentially the pipe must function well. The foreign exchange market and secondary corporate bond markets are also nothing to write home about either. Persistent levels of excess liquidity and the phenomenon of many banks chasing a concentrated set of corporate players (and these days MIC convertible bond pricing is madness) has distorted credit risk pricing and biased the capital structure of non-bank corporations over the years.
A formal flexible inflation targeting regime
So what should the Bank of Mauritius do now as it works on implementing a new monetary policy framework?
Firstly, the Bank of Mauritius should look at adopting a formal flexible inflation targeting regime which would suit a small and open economy well. It should focus on either core inflation or year-on-year inflation rather than some moving average headline inflation rate. When it comes to picking an actual target, it should first work with Statistics Mauritius in order to update the consumer basket taking pandemic effects into account (things will not be like before) and look at how different income groups across the distribution (versus the mean or median) face different inflation rates over time. It should also look at the inflation rates of its export competitors and export targets.
Costly bailouts occur at the same time as banks are making near pre-pandemic profits.
Mauritius is a small and open economy and while competing objectives can weaken the transmission mechanism of monetary policy, a flexible target where a +/- 100 basis points band is set around say a 3% inflation target could provide some flexibility to the central bank. The band could even be as wide as 150 basis points initially (but no more, else one would lose what is known as the honeymoon effect) with a focus on reducing it towards 50 basis points over the longer term. The central bank would clearly announce its medium term forecasts to the market within the constraints of the band in order to guide market expectations while allowing for some leg room for other considerations.
Secondly, the central bank will need to strengthen its balance sheet in order to become more credible and depend less on unrealized rupee depreciation gains from international reserves whose value in rupees is technically meaningless. Rupee depreciation after all leads to higher prices, and such shocks may last longer than expected given our import dependence and the size of the economy.
The priority for the Bank of Mauritius right now would be to immediately conduct a holistic asset liability study taking into account all liabilities, current and expected. In many ways monetary policy is a dynamic process, and commercial bank balances and assets held at the central bank will fluctuate over time. Models have limitations and scenario analysis will be key here.
The Bank of Mauritius must also account for other liabilities such as its own staff pension liabilities (along with the most appropriate discount rate) and risks associated with the non payment of part of the MUR 60 billion grant/loan (which one is it anyway?). On the asset side, it would need to come up with expected return projections (including their distributions and expected risk measures) for both international asset classes and MIC investments.
When it comes to the MIC, this was one of the reasons why this author was very vocal about not having such things on the central bank balance sheet and certainly why this author wrote extensively about why the mis-pricing of the “convertible” bonds would lead to mark to market losses from day one if marked to market (as convertibles should be!). In essence, the Bank of Mauritius will need to hold overvalued bonds on its balance sheet until maturity (assuming no default) in order to make unspectacular returns for the risk taken.
This is, as I explained in previous articles with the math to back it up, because even if some stock prices have rallied recently albeit on thin volume, the European style of the option when coupled with the issuer’s ability to call back the bonds at anytime essentially kills the option value of these bonds. These bonds are structured in such a way that they heavily favour the issuer. Yes, Covid came but that does not mean that we needed to go so far especially when taking the balance sheet implications into account.
The private sector may be keen to call these bonds “quasi-equity” and show that they are de-leveraging on paper (some historically troubled companies are also doing quite the few fair value adjustments on unrealized gains and even some pension discount rate adjustments). But when a convertible has no option value, it behaves like a plain vanilla bond and should be valued as such and treated as such (debt). The less said about why some players wish to call it “quasi equity” despite the math, the better.
In a country where costly bailouts which will have longer term balance sheet and monetary policy implications occur at the same time as banks are making near pre-pandemic profits, one has to wonder about the need for such a large MIC portfolio, and one should think about having a more balanced compensation structure between commercial banks whose balance sheets have been partially cleansed and the central bank which has been the only central bank in the world to take on all this risk directly on its balance sheet.
The more domestic assets under-deliver, the more active the Bank of Mauritius will need to be in global markets.
Strategic and tactical asset allocations
Getting back to the Bank of Mauritius balance sheet, it would essentially need to do quite the sophisticated asset liability management study in order to design an appropriate strategic asset allocation framework for international reserves which would not only meet the country’s external liabilities but also central bank’s own liabilities. With bond yields globally at near historical lows and inflation expected to remain moderately above its 20-year historical average in the coming decade, the worst thing one can do is go for duration. Given its liabilities, the Bank of Mauritius will need a strategic asset allocation which includes global equities, liquid alternatives and, indeed, less liquid asset classes as well. However, a strategic asset allocation is not useful unless you actively manage risk factors dynamically at the total portfolio level.
The Bank of Mauritius must strengthen its investment team to include experts in factor investing which would include derivatives overlays given that a strong tactical asset allocation framework will also be needed given its balance sheet liability profile. In Mauritius, fund managers typically do simple fund of funds in the private sector without a proper understanding of how risk factors interact with each other and change. All this needs to be actively managed at the mother portfolio level (taking all positions into account at macro level). Fund of fund managers globally do more advanced things than what is done in Mauritius by our local fund managers. Size likely matters.
In this respect, beyond strengthening its core investment team, the Bank of Mauritius should also seek external advisors and ensure that portfolio managers have relevant experience including solid quantitative experience in managing sophisticated multi asset multi strategy portfolios from a factor lens. The Bank of Mauritius must also ensure that it can preserve and grow the real purchasing power of its international reserves over time, which essentially means that as risk factors and risk levels change on a daily basis (more like intra-day), it must be able to manage ex ante tail risk.
Now more than ever, active portfolio management will be key. Tail risk hedge implementation is complex and requires having the right people at the right places. Remember, in global markets, only the best survives. Bears make money, bulls make money, but pigs get slaughtered. In many ways the strategic asset allocation and tactical asset allocation for international reserves will need to be a subset of a more holistic allocation of both domestic and foreign assets.
The Bank of Mauritius must be clear to the MIC in terms of derived return expectations on future investments. The more domestic assets under-deliver, the more active it will need to be in global markets, which will require a review of how much risk its Board is willing to take. This is very complicated to understand unless one comes from this field. These are very complex and quantitatively driven discussions coupled with multiple scenario analyses requiring a very sophisticated and disciplined risk management framework and appreciation of institutional investing.
Central bank digital currencies
Thirdly, fixing a fragmented secondary bond market would involve multiple government bond buybacks spread over years in order to reduce rupee bond market fragmentation and essentially build a better pipe. While such efforts should be pursued, the Bank of Mauritius appears to be correctly thinking out of the box here, and the current Governor should be given credit for that. While we do not yet have any details on this new monetary policy framework, one would clearly assume that it would involve central bank digital currencies (CBDC).
We can define a central bank digital currency simply as an electronic, fiat claim on a central bank that can be used to settle payments or as a store of value. It is in essence electronic central bank, or narrow money. Conceptually, a central bank digital currency can simply be viewed as an electronic, fiat claim on a central bank that can be used to settle payments or as a store of value. It is in essence electronic central bank, or narrow money.
Central Bank Digital Currencies over time would also become an effective liquidity management tool.
In the case of Mauritius, the Bank of Mauritius’ CBDC would essentially be a risk-free proxy and offer interest which would, along with the traditional interest on reserve, help set a floor for short term interest rates. The CBDC rate would be accessible to anyone, not just commercial banks however. Over time banks would need to offer interest rates that are slightly to moderately above the floor given that holders of commercial bank deposits could quickly switch to CBDCs otherwise.
The central bank would have a flexible inflation target. It would set a level of interest rate which would be consistent with closing the output gap over the medium term and reaching its quantified inflation objective (within the band) and beyond the typical reverse repo transactions. It would also set the CBDC rate very close to its policy rate (adjusted for the convenience of ease of transactions).
CBDCs over time would also become an effective liquidity management tool as they would become part of the central bank’s liabilities. Liabilities get into the balance sheet complications discussed earlier. The central bank typically sets a policy rate with a floor as described above, allowing it to take liquidity out of the system, and a ceiling allowing it to inject money into the system. Essentially, the central bank in a CBDC world not only can lend money to commercial banks at the short end of the maturity spectrum (also allowing it to influence rates commercial banks set on deposits as well as what the floor does), but it can also lend CBDCs to any player and, in theory, across the maturity spectrum.
This opens up new policy avenues and risks. The best path forward in the current context beyond lending to commercial banks via highly collateralized setups (and in the context of keeping the term premia of the yield curve low and long term borrowers including the government happy) would be for the central bank to also purchase longer duration government bonds and earn the term premia while expanding its assets (and equity capital). CBDC interest bearing liabilities do need to be financed somehow after all.
Such a Quantitative Easing like programme, subject and constrained by a central bank’s inflation stance at time T, should be spread over a long period with controlled amounts so as not to distort bond markets. Subject to its inflation target, the central bank could also, when warranted, lend CBDCs to over-collateralized notes or funds which would, at their level independently, lend money to the real economy. Note here that the Bank of Mauritius would not be the direct lender but would only lend to a special purpose vehicle (SPV). Credit risk for the Bank of Mauritius would be managed via over-collateralization and by the diversification strategy followed by the SPV. Such SPVs could, for example, be focusing on ESG initiatives or any target sector where monetary policy makers would want to focus on.
The border between supply side and demand side inflation may become blurry very quickly.
The ability of the central bank to influence the Prime Lending Rate would be significantly enhanced. All this said, the Bank of Mauritius should follow a very gradualist approach because CBDCs as part of a flexible inflation targeting framework is new age stuff. There are obvious implications for commercial banks as well. Initially, the most likely structure is for both the floor and ceiling to focus at the short end of the maturity spectrum. It is just that while the secondary bond market gets developed (the traditional pipe), CBDCs open an interesting set of possibilities and risks.
All of this requires quite the sophisticated setup, especially when it comes to how the Bank of Mauritius will need to manage risks, optimize return on assets and in general, strengthen its balance sheet. While one can hope for government funded re-capitalization spread over years, the Bank of Mauritius will need to quickly establish its monetary policy framework in order to more firmly anchor inflation expectations beyond the current transitory supply side driven inflation surge. As the economy recovers more strongly given the opening of borders, the velocity of money may also increase more rapidly than anticipated. Already, the civil servants want more money and the private sector employees do too. The border between transitory and stubborn, the border between supply side and demand side inflation, may become blurry very quickly.