Abheek Barua: Two tales of usury

The story of Indian interest rates, as it unfolds now, seems to have two somewhat distinct sub-plots. The first clearly revolves around the uptick in the business cycle and the consequent pressures on inflation, demand for funds et al.

The second, more insidious, strand revolves around a process of correction in markets such as housing finance, where intense competition had thrown rates completely out of line with the risk profile of borrowers.

The denouement of the story, as the two plots come together, is likely to be higher borrowing costs for segments like retail lending and more sustainable margins for banks.

The story of the cyclical upswing and its impact on interest rates is fairly well known. After a year and a half of sustained recovery in the manufacturing sector, companies find themselves with very high levels of asset utilisation and are beginning to expand capacity.

These capacities are being funded through a combination of internal accruals, bank borrowings, and, to a limited extent, equity IPOs. The increase in the demand for funds is pressuring up interest rates. Couple this with a sharp rise in domestic inflation and tightening global liquidity and you have a somewhat clear case for a rise in domestic rates.

How high are rates likely to go? To a certain extent, it depends on what one refers to as "interest rate", specifically the tenor of the loan for which the rate is defined.

During a phase of recovery in an economy, long-tenor rates (read this as five years or more in the Indian context) tend to rise faster than short-duration rates. Over the last three months, Indian rate changes have stuck to this formula.

Average yields on debt securities of the ten-year tenor have moved up by a good 120 basis points (a basis point is a hundredth of a percentage point) while one-year yields have moved up by roughly a third of that. This is likely to be the pattern of interest rate movements over the next few months""long-term money will become dearer than short-term money.

To get back to the core issue of how much more yields or interest rates are likely to go up, I would argue that they are unlikely to go up much more. The doomsdayers have it that inflation numbers continue to move northward.

I would tend to argue that the debt market has priced in quite a bit of these increases (that stems essentially from higher fuel and metals prices) and have factored it into yields long before the macro inflation numbers were actually announced.

Rate hikes by the US Fed of roughly 75""100 basis points have also been priced into current yields. This is not to deny that some more increase is likely. But I would be really surprised if ten-year yields cross the 7 per cent mark and continue upwards.

What about the RBI's likely response to rising inflation rates? Isn't it the central bank's remit to smother inflationary pressures by tightening the monetary levers? Here it is important to identify and distinguish between the sources of rising inflation.

If inflation is driven primarily by rising levels of demand that easier monetary policy will exacerbate, then there is a clear case for hiking policy-driven rates (like the repo rate or the bank rate).

If, on the other hand, the impetus comes largely from discrete, identifiable cost "pushes" like fuel price increases, the case for contractionary policy action becomes weaker. Particularly so, when investment activity in the economy shows signs of picking up.

If the RBI does try to push rates up, it could potentially land us in "stagflation", where despite higher interest rates, prices keep rising on the back of fuel price hikes while simultaneously, economic growth and investments are stifled by rising rates.

A more sensible strategy would be to address the price increases in the commodity sectors directly through a combination of lower tariffs, administrative intervention (as in the case of oil), and moral "suasion" of commodity producers.

Let me turn to the second sub-plot in the story that is pegged on housing finance giant HDFC's decision to hike housing loan rates a couple of weeks back.

Most of my thesis here is incidentally based on my Crisil colleague Geeta Chugh's recent work ("Low incremental return on housing finance to harden rates," Insight, vol I, no 33).

The HDFC decision is being seen as part of the larger cyclical story where a rise in government bond yields is likely to translate into rises in other rates as well.

While that is true to a certain extent, the HDFC decision also reveals a key structural problem in the banking industry. Chronic "excess capacity" in the banking industry and intense competition led to a complete mispricing of loans in high-growth segments like housing loans.

Between April 2000 and February 2004, the rate incidentally dropped by 675 basis points, compared to a decline of 580 basis points in 10-year government securities. This warranted immediate correction and the HDFC decision is a first step in bringing it about.

Chugh points out in an effort to grab a piece of the action, banks drove borrowing rates so low last year that the return on equity (RoE) of the housing finance (henceforth HF) business plunged to 9.01 per cent from 18.8 per cent in 2002-03.

On the other hand, contrary to the myth that the housing finance sector is low-risk and associated with low non-performing loans (NPLs), NPL ratios in the sector have actually been rising quite sharply.

Apart from genuine economic problems such as the loss of jobs and life, it is also due to a growing population of willing defaulters and fraudsters, particularly over the last two years.

Using lagged NPLs (on March 31, 2003, as a proportion of outstanding housing loans as outstanding loans as on March 31, 2002), the Crisil study finds the proportion of NPLs to loans for the HF segment as a whole was 3.3 per cent. This is way above the NPL-loan ratio in a developed market like the US, where the comparable number is just 0.78. The high level of NPLs naturally entails high provisions and depresses margins. To correct for this, a rise in lending rates is imperative.

Besides the problem of rising NPLs, the lack of comprehensive data (like a credit bureau) on retail borrowers has its problems. Bankers point out that the current rates charged on retail loans do not cover the costs of verifying the risk profile of consumers. The bottom line, again, is that if retail lending has to be viable, the price of loans has to increase.

The prolonged period of easy liquidity over the last two years has led to links in the risk-return structure that need to be ironed out. Ironing out in this case implies a rise in lending rates and bankers are likely to use the impetus given by the cyclical pressures for rate increases to push through this correction.

(The author is Senior Economist at Crisil. The views expressed are personal)