In early March, alarmed by the flood of money rushing into small-cap funds, after a massive one-year bull run, the market regulator asked mutual funds to consider moderating flows and rebalancing portfolios, since the flows could make the market frothier. The Securities and Exchange Board of India (SEBI) also asked mutual funds to conduct a 'stress test' for small and mid-cap schemes and publish the results. The stress test covers liquidity, volatility, valuation and portfolio turnover in such schemes, “along with guidance in simple language, assumptions and methodology, to enable the investor to understand the risk associated.” The core of the stress test is this: the number of days required to liquidate 25% and 50% of small-cap and mid-cap portfolios.
Regulating the flows would, indeed, amount to directly controlling the frothiness of the market (it is money that mainly drives stocks; money, in turn, is driven by sentiment and earnings), but the stress test is another matter. There are four problems with the regulator-mandated stress test. Firstly, at the heart of stress tests is a fallacy: it is assumed that the average volume in a frothy, bullish market will continue in a sharply falling market. But the point about markets is that liquidity can expand and contract dramatically. In a rising market, there are plenty of buyers and sellers. Under extreme stress, there are no buyers, only desperate sellers. Volumes simply disappear. How many days it would take to liquidate 50% or 25% of the portfolio will vary enormously, depending on market climate. No wonder, a veteran portfolio manager has called the stress test results ‘useless’. Stress tests do not take into account what is technically called non-linearities and fat tails, or extreme situations which cause amplification effects, leading to chaos, confusion, contagion and impulsive decisions, even by experienced market participants. It is similar to the scene when someone shouts 'fire' in a crowded theatre.
Secondly, what is an investor going to do with a bunch of disparate metrics like the number of days to liquidate a portfolio or concentration (in large-cap, mid-cap and small-cap), standard deviation, beta, portfolio turnover and price-to-earnings ratio? How will she decide to switch from the one with adverse metrics to a better one? What if a scheme scores high in some parameters and lower in others? For example, a scheme may have low-valuation stocks (which may decline less in a falling market) but these stocks take longer to liquidate; how will an investor weigh the merit of the first with the drawback of the second? She will then have to weigh other metrics like standard deviation, portfolio beta and turnover ratio. Doesn’t someone need to assign weights to all these factors and put all the schemes on a common footing, calculate the total score and then create a ranking? If this sounds like too much work, is there any other logical way to use these factors? If I can indeed rank schemes based on the results of a stress test, what action will I take? Will a higher-scoring scheme lead to a higher risk-adjusted return? How will I know? Has anyone back-tested how useful these stress test parameters are for making better investment decisions? Assuming we focus only on one important factor—the number of days that it takes to liquidate a portfolio—what should I do, if I am a long-term investor in a systematic investment plan (SIP)? SBI Mutual Fund, which has 82% in small-cap stocks, will take 58 days to liquidate 50% of its portfolio but UTI Small Cap Fund, which has 83% in small-caps, will take just five days. The range of days is too wide to be meaningful (Source: Association of Mutual Funds of India website).
Thirdly, SEBI’s definition of small-cap stocks is a market value of less than Rs5,000 crore. Many sectors and many stocks have participated in the ongoing bull market. A quick check on screener.in tells me that there are 3,898 companies with lower than Rs5,000 core market value and only 688 companies are higher. This means that 85% of the market is classified as small-cap. This is a one-size-fits-all approach that will inevitably lead to weird outcomes. A scrip with a market value of Rs4,999 crore is small-cap, as is a company with quarter the value. They will have different liquidity and volatility profiles, but will be classified under the same small-cap label. I believe that the current three-class division (large, mid and small) is inadequate, so we need to classify stocks under at least five categories.
Fourthly, the stress tests put too much weightage on factors that are not relevant to the fund manager. There are multiple approaches to stock-picking but the most popular one is picking stocks with long-term prospects available at a reasonable valuation. These are usually found among small- and mid-caps. It is not easy to find such stocks, especially after a strong bull market. Will a fund manager ignore such a stock if its liquidity is low? Certainly not. But today, such stock picks will hurt the liquidity criteria under the stress test. For all these reasons, the whole idea of stress tests rests on shaky ground. It would go the same way as the 'risk-o-meter' disclosed along with the schemes—well-meaning but not very practical.
(This article first appeared in Business Standard newspaper)