Enhancing effectiveness of monetary policy in Bangladesh
Mustafa K Mujeri | Published: January 22, 2021 20:32:43
Since the 1997 Asian financial crisis and the subsequent 2007-2008 global financial crisis, the global economy has entered into a period of elevated economic volatility, underpinning the need to use more effective policy tools for macroeconomic stabilisation, especially for the short term. From such compulsions, the importance of monetary policy has significantly increased in the absence of more appropriate tools. At the same time, there exists heightened scepticism regarding the efficacy of monetary policy in developing economies such as Bangladesh, although monetary policy tools play a central role in short term stabilisation in developed economies. Hence, a natural question for the policy makers in Bangladesh is: how effective is monetary policy in Bangladesh?
For the Bangladesh Bank (BB) to achieve its mandate, it is important that monetary transmission works seamlessly-a process through which the BB’s policy action is transmitted to the ultimate objective of stable inflation and sustainable high growth. The policy action consists typically of changing the interest rate at which it borrows or lends ‘reserves’ on an overnight basis with the commercial banks. The transmission mechanism hinges crucially on how monetary policy changes influence households’ and enterprises’ behaviour. This change can take place through several channels. The primary channels are: (i) interest rate channel; (ii) credit (bank lending) channel; (iii) exchange rate channel; and (iv) asset price channel.
Here, we take a particular monetary policy transmission channel as a basis for measuring monetary policy effectiveness in Bangladesh-the bank lending channel. This relates as to how changes in the monetary policy instruments by BB affect the lending behaviour of the banks. More specifically, BB’s monetary policy will be considered more effective if a policy to trigger monetary expansion can elicit a larger reduction in the lending rates of the banks, thus lowering the financing cost for the businesses. And lower financing cost may be expected to encourage lending and stimulate investment and economic activities.
The immediate impact of a change in the monetary policy rate is on short-term money market rates. A bank will be willing to part with its reserves overnight to another bank only if it earns at least the rate that it could earn by keeping these funds with BB; and, if banks compete adequately for such lending, then the rate will in fact track closely the BB’s policy rate.
In turn, BB’s policy rate change will impact the banks’ cost of funds, both the rates they would pay to the depositors and the rates they would demand for making loans. For example, when BB reduces the policy repo rate with the intention to support aggregate demand in the economy, the expectation is that there would be a reduction in the banks’ cost of funds and lending rates, and other market interest rates. Lower lending interest rates of banks provide a boost to the demand for bank credit from various segments of society, and hence, increases in overall demand, income, and output in the economy.
The implicit assumption in the above is that bank balance sheets are strong and in a position to step-up quickly the supply of credit in response to lower funding cost and higher demand for credit – the bank lending or the credit channel of transmission. However, monetary transmission is greatly hindered if bank balance sheets are weak and they do not have much loss-absorption capacity to deal with their problem loans, resulting in accumulation of bad loans, misallocation of resources, productivity losses and slow growth. This way, attempts to stimulate growth with aggressive policy rate cuts when there are bank balance-sheet problems may not work and may lead to deeper balance sheet crisis.
Since Bangladesh’s financial system is overwhelmingly bank-dominated, the overall efficacy of monetary transmission depends critically on the extent and the pace with which banks adjust their deposit and lending rates and meet adequately the economy’s demand for credit due to changes in the policy repo rate. Overall, empirical evidence suggests that the pass-through from policy rate changes to bank lending rates is slow and subdued in Bangladesh. This lack of adequate monetary transmission is a key policy concern for the BB.
In Bangladesh, a large proportion of the loans is at floating rates i.e., the interest rate charged to the borrower keeps changing depending on the reset periodicity. The floating rate is linked to some ‘benchmark rate’. Banks also charge a spread over the benchmark to factor-in term premium and credit risk, among other factors. The actual lending rate is the benchmark plus the spread. The benchmark could be internal or external; an internal benchmark is based on elements which are in part under the control of the bank such as cost of funds, while an external benchmark is outside the control of the bank (for example, it could be market determined rate such as certificate of deposit rate or treasury bill rate or inter-bank offer rate, or it could simply be the central bank’s policy rate).
The virtue of an external benchmark is that it is transparent, common across banks, and borrowers can compare various loan offers by simply comparing spreads over the benchmark. As market rates normally move in line with the central bank’s policy rate, an external benchmark is globally considered and adopted as more appropriate than an internal benchmark for transmitting monetary policy signals. In Bangladesh, banks enjoy the flexibility to use both the internal and external benchmarks, but the banks seem to prefer internal benchmarks on two counts: first, the internal benchmark reflects their cost of funds; and second, it is perceived that there do not exist any robust and vibrant external benchmarks.
In this context, what is important for BB is to explore possibilities to reduce the weaknesses and rigidities in the transmission mechanism through effective measures. For example, the banks may determine their benchmark lending rates taking into account the marginal cost of funds, such that the lending rates could be more sensitive to changes in the policy rate. The actual lending rate may also include a spread for covering the costs of business strategy, credit risk premium and others. This will give better transparency, more flexibility and faster transmission.
The low pass-through from policy rate to bank lending rates is affected either by the banks not changing the benchmark rate or by adjusting the spread. The rate rigidity on the liability side of the banks may be caused by several factors in Bangladesh. In the country’s banking sector, the overwhelming share of total liabilities of banks is in the form of deposits. Bank deposits are predominantly at fixed interest rates, thereby imparting rigidity to the transmission process. Further, more than one-third of the deposit accounts are fixed deposits, implying that their rates get reset infrequently and with significant lags to policy rate changes. While the banks’ marginal cost of funds may drop quickly with a cut in fresh deposit rates, the average cost of deposits comes down rather slowly, which weakens the transmission.
Further, the banks in Bangladesh have a large access to low cost current and savings account funds. These funds constitute about 30 per cent of aggregate bank deposits with the share of saving deposits at around 20 per cent. More importantly, the banks are mostly free to decide saving deposit interest rates which do not necessarily follow the monetary policy signals. In addition, the deterioration in the banking sector’s health due to worsening of asset quality over the past several years and the expected loan losses in credit portfolios also induce large variability in spreads in the pricing of assets. Indeed, the weak banks aim to maintain profitability in the short-term even at the expense of long-term profits as well as deposits and lending shares. This also has significant impact on the transmission to the lending rates. These and other factors contribute towards rigidity in the liability side of banks’ balance sheet with respect to policy rate changes, inducing a behaviour that makes the rates on the asset side of banks’ balance sheet rigid as well.
Bangladesh Bank needs to take steps to enhance the transparency and transmission from monetary policy signals to the actual lending rates of the banks. In the long run, the need is to shift to an external benchmark based lending rate system as the internal benchmark based pricing regimes are not consistent with global practices on pricing of bank loans. The transformation should be pursued in a time-bound manner. While recognising that no external instrument in Bangladesh meets all the requirements of an ideal benchmark, the Treasury Bill rate or the BB’s policy repo rate may better serve as an external benchmark. Similarly, the periodicity of resetting the interest rates by banks on all floating rate loans, retail as well as corporate, may be reduced to once in a quarter to expedite the pass through from the monetary policy signal to the actual lending rates. To reduce rigidity on the liabilities side, the banks may be encouraged to accept deposits, especially bulk deposits, at floating rates linked directly to the selected external benchmark.
Efforts should be directed to improve the monetary transmission by ensuring that changes in the policy rate transmit quickly and adequately to banks’ lending rates in a transparent manner. The aim is to make the banks’ liability side more flexible so that the objectives of improving monetary transmission by BB and maintaining healthy net interest rate margins by the banks are effectively aligned.
Obviously, efficient monetary transmission is a pre-requisite for the successful pursuit of its objectives by any central bank. The transmission from the policy repo rate to bank lending rates, which is the dominant transmission channel in Bangladesh, remains an issue of concern. The key issue is to address: who should bear the interest rate risk in the economy – the borrower, or the depositor, or the bank? Retail depositors and borrowers are unlikely to have efficient tools to manage the interest rate risk. Banks, however, should have the ability to manage the interest rate risk. Similarly, large depositors and large corporate borrowers can also be expected to be in a position to manage the interest rate risk. Non-bank financial institutions with less exposure to interest rate risk, such as insurance and pension funds, could also be good repositories of this risk. A combination of interest-rate risk transfer mechanisms through market products such as interest rate derivatives (e.g. swaps) and securitised products such as collateralised loan obligations (CLOs) can also be good vehicles.
In particular, monetary policy will be more effective with deeper financial markets, more competitive banking system, stronger institutional and regulatory environment, and a more transparent BB. On the other hand, higher integration with the international financial markets with greater exposure to global shocks will limit the role of monetary policy and dampen its effect.
Going forward, monetary policy will have to play a bigger role in Bangladesh in the coming years. For the purpose, BB needs to adopt new instruments and procedures, improve its existing frameworks, and gain greater autonomy in the conduct of monetary policy. The government, on its part, will have to strengthen its efforts to enhance BB’s transparency and refine policy and operational frameworks. Structural reforms are also necessary to strengthen institutional and regulatory structures, allow greater competition in the banking sector, and deepen financial markets.
It is well known that monetary policy changes are more likely to affect the bank lending rates in the theoretically expected directions in countries that have better institutional frameworks, more developed financial structures, and less concentrated banking systems. Bangladesh scores poorly in all of these dimensions, and hence the country indeed exhibits much weaker transmission of monetary policy changes to bank lending rates than do the advanced economies.