Policy transmission: Some fallacies

Across the world, the issue of transmission of monetary policy impulses (through the signalling rate) specifically through the interest rate channel warrants a debate. This is all the more relevant in the Indian context where there is a serious lack of understanding on this particular issue among various stakeholders. The moral of the story: Banks are often blamed for incomplete monetary policy transmissions! Let us understand what really is the economics behind it.

An understanding of how the policy transmission works is best deciphered from the asset-liability structure of banks. A cross-country analysis of asset and liability structure suggests that deposits from public (primarily demand/CASA and time deposits) mostly fund the investment and advances on the asset side for Asian economies such as India, Bangladesh, Philippines, and even Japan. For countries such as Singapore, borrowings are a significant percentage of liabilities. Interestingly even for countries such as the UK and the US, the share of deposits is much less at about 65-70 per cent, and that too with a caveat. 

The large share of public deposits in total liabilities for countries such as India has important implications for macro stability and policy transmission. Firstly, with banks funding themselves through retail deposits, the source of vulnerability to external contagion is significantly reduced. Second, only 1 per cent of the bank borrowings are currently at the policy rate of 6 per cent. Third, the share of public deposits has a preponderance of CASA (41 per cent approximately) that is mostly interest rate agnostic in India with an average interest rate of about 3.5 per cent. The rest are time deposits with a fixed interest rate for the duration of the deposit tenure. Thus, when say repo rate changes by 25 basis points, even under full transmission there could be at most a 15 basis point impact on deposit rates (25 bpx59 per cent interest-sensitive time deposits) and thereby on lending rates. 

Now let us compare India with other countries to put such transmission in proper arithmetic. In developed countries, the financial system operates primarily in a liquidity shortage mode (as deposits are non sticky) and one has to take frequent recourse to central bank liquidity. Indian banks, on the other hand, are mostly deposit-driven that are sticky and do not resort to borrowings on a larger scale.

Next, let us take the examples of individual countries. First, take the example of Japan that has overall deposit liability structure similar to India. The most popular private home loans in Japan are a combination of interest rate loans, with an initial five-, seven- or 10-year fixed period, and variable interest rate during the remainder of the term (usually 20-30 years). Ten-year fixed rate home loans closely follow movements in the 10-year government bond yield, while variable rate home loans are reset every six months.

Most residential mortgages in US have long-term fixed rates, particularly 30 years and 15 years, but almost all pre-pay due to sale or re-financing. Financial institutions bundle mortgage loans into securities that are traded in a secondary market. The purpose of this system is to quickly free up money (given the nature of the system) for the financial institutions to lend out in the form of new mortgages. Such residential mortgages are priced from US 10-year treasuries with a concomitant risk spread.

In the UK, mortgage loan financing relies less on securitised assets than the US and more on deposits such as Australia and Spain. Thus, lenders prefer mostly variable-rate mortgages to fixed-rate mortgages to reduce potential interest rate risks. There is a common argument thread across all such cases. First, contrary to India, demand deposits in all such countries literally do not pay any interest rate and can be withdrawn at any point of time without any penalty. This minimises the cost of deposits significantly.

Second, even in the time deposit category, the deposits are mostly floating and are linked to the bank’s external bench mark. Again, such mechanism minimises the cost significantly. 

In India, banks had taken a first stab at floating deposit rates back in 2001 and even in 2010 with rates even linked to yields on government securities.  However, these schemes received a thumbs down from customers. Such a response was not entirely unexpected, since India has a limited social security system in place and senior citizens rely on bank deposits as a source of retirement corpus. In contrast, senior citizens in developed countries are entitled to generous social security corpus and need not rely on bank deposits. 

Third, the external benchmark is popular in western countries in setting lending rates as banks there depend largely on short-term flexible funds. This condition is unlikely to ever emerge in India.

What do we decipher from the above mentioned analysis? There is hence no denying the fact that it is only an established global practice of bank liabilities being the ultimate deciding factor in pricing of the assets. However such an arrangement typically implies a global practice of floating or even neglible interest rate on bank deposits and reliance more on short term liquidity through borrowings. This gives such banks the operational freedom to price assets at market rates. In a developing country like India that has limited social security system in place are we ready to price all liabilities like even Government sponsored small savings at market rates? Are we aware that share of only bank deposits in total financial savings is today still at 40 per cent? Can we nudge risk-averse retired people to move away from bank deposits by guaranteeing them social security?

For the record, Indian banks currently have one of the lowest NIM at 2.8 per cent (world average 5.7 per cent). Can it go further down? Let us answer all these questions before we perhaps put a stop to this transmission debate!
 

The writer is group chief economic advisor, State Bank of India. Views are personal