Monday 28 October 2013

The Risk Based Capital framework in the Indian context

Mr. J. Hari Narayan (JHN) the immediate past chairman of the Insurance Regulatory & Development Authority of India (IRDA) in conversation with Abraham Kuruvilla (AK) on issues relating to adoption of a ‘risk based capital’ framework by the Indian insurance industry.

AK : Hello Mr Hari Narayan, how important is it for
the Indian Insurance industry to adopt a risk-based capital framework and are they ready for it?

JHN: There is a risk-based capital concept in India, but it is not within the terms of what is understood in international parlance. And particularly after the Solvency-II directive has come out! That’s the first thing we must understand. Now, within the risk-based solvency framework that they are talking about, companies are expected to assess their own risk, and create an internal model as they call it, to assess the risk which they have on their liabilities side, as well as their assets side.

Now they are entitled to use their own model and calculate their risk based upon that particular model. The model however is subject to oversight by the regulator. In the event a company doesn’t choose to develop its own model, or if the model developed is inadequate, then the regulator may prescribe a model, which the company is bound to follow. So fundamentally, that is the pattern they follow – or rather what they have adopted within the terms of Solvency–II. Now in India, we have a different concept of solvency and we believe that our system is far more robust

AK :: Related to the problems that Europe is
having with Solvency–II, what do you foresee as some of the problems that India would have in adopting the framework?

JHN: See, merely because it is adopted in Europe, that doesn’t mean it’s the best model to go by. Having said that, the need for solvency is a cornerstone of the financial structure of an insurance company, and it is so across the entire world. Briefly, what it says is that at any point of time, the insurance company must have assets which are sufficient to meet its liabilities. That law applies not just in Europe, it applies all over the world and in India too, and it is always applied in India. The question now is, how are you going to measure your assets, and how do you measure your liabilities. And how are you ensuring that one equals the other. That’s the question!

So when they are talking about risk and risk-based assessments, they’re really talking about the accuracy of estimates. For an insurance company, whether it be liabilities or assets, both are only estimates. No doubt competent estimates, professionally executed estimates, but estimates nevertheless. Take for example the question of estimating liabilities; typically, insurance companies look at these liabilities based largely on the past data and past track record, and project the liabilities in the future. Now, this might work well, but I think you know about the ‘Black Swan’ event don’t you?

AK : Yes, I’ve some idea!

It is a term which has become very popular after the 2008 financial crisis –fundamentally, the question is: how are you going to, based on past experience project, for instance, a disaster like a Fukushima, or the Christchurch earthquake, or even the Kedarnath landslide and floods and all that? Disasters of this dimension are very difficult to predict. That’s why the estimate of liabilities is always shaky. It is founded, no doubt, on statistical mathematics, but fundamentally, since it’s based upon the past, is that a very reliable indicator to the future or not, is a debate that is going on at a very esoteric level in the world of mathematics.

But having said that, essentially when you’re saying ‘risk-based’, what you’re saying is: a) we need to assess what might be our liabilities at a point of time, and b) the same thing holds with regard to assets – what is the value of assets which an insurance company holds. If you take insurance companies in India, for instance – and that’s largely true across the world, the pattern is not very dissimilar – in India, for instance, insurance companies hold about 15% of their assets in equities. And the other 85%, bulk of it – about 55% to 60% - is held in government securities, and the balance 25% to 30% is held in corporate debt. So you’ve got government debt, corporate debt and equities. As far as equities are concerned, there is no problem, because on the date when you are going to assess, the value of a given asset, there is a market value because there is discovery in most of these shares. That is why in India, for instance, we do not allow insurance companies to invest in the equities of non-listed companies. The reason is, there is not sufficient discovery of price in such instruments. But that in any case amounts to only 15% of the portfolio, generally. So the question is: how are you going to assess the debt? What is the value of the debt which one has? And that is the issue, as far as India is concerned.

AK : Could you elaborate?

JHN: There is a myth that all government securities are risk-free. Now we have seen in the 2008 crisis that that may or may not be a valid proposition. There are several government debts, for example, which certainly were not risk-free. So the question is whether it is wise to look at Indian government debt as risk-free. And even government debt in India is of two kinds; one is, you’ve got the central government debt, there are also state government debts, and there is also the debt, or the bonds issued by state government instrumentalities, like for example Power Finance Corporation and other corporations and so on. So what is the value of this debt? Is it risk-free, truly?

And then there is the question of corporate debt, and where exactly the corporate debt is. If you look at the banks now, banks have lent money to very big corporates, and several of them are in distress, because of various reasons, the repayment of the loans taken by the corporates is not as per schedule, which is why they’ve got a problem of the asset quality, and they have to make provisions for asset quality. So when we’re looking at the portfolio of an insurance company’s assets, the question is: what provisions do we need to make for different classes of debt? That is the question. And we don’t have a sufficiently robust and deep statistical basis on which one can assess that the risk in a given instrument is so much. The banks have done it.

AK : So, Sir how do we do it?

JHN: The system we follow in India is, that the liabilities side can be fairly well-assessed, because as far as life insurance companies are concerned, there’s little problem in estimating liabilities, because the statistical basis of life and life probabilities and events and so on is very well-charted, and you can do it very accurately. The same doesn’t hold good for non-life companies, that’s general insurance companies, which deals with miscellaneous types of perils, like for example, as I mentioned, the Uttarkashi issue, or the Bombay flooding and things like this. But in India, fortunately, those secular kinds of perils are not very widely insured. The bulk of our insurance is in motor car insurance and in health insurance, and these are very well tracked. So there’s not much of a problem. The problem comes only when you’ve really got vast property holdings, and then the issue of assessment of risk is of far greater importance.

So coming back, in India, what we do – we require insurance companies at all times to maintain one and a half times the assets over the liabilities as solvencies. In other words, it is not equal to, but it is one and a half times greater than. And that’s how I believe it’s a very robust thing. But on the downside, because we haven’t got well developed markers to assess what might be the quality inherent in, let us say, corporate debt, or even in a government debt, we’ve hiked up the percentage, partly to meet the inaccuracy in estimating the risk on that factor, and secondly it’s also the question of time.

Supposing we find that a company’s assets are eroding, or their liabilities are increasing over their assets, it takes some time before that can be recognized. Their system of accounting has to come up from that, it has to be recognized, it has to be reported, and so on. And having recognized it, we will again require time – of both the management and the regulator – to take sufficient steps to restore the equilibrium of the company. That also takes time. And in order to allow for this time, as also allowing for a certain uncertainty in the quality of the debt, we’ve insisted on a margin of a 150%. So that gives us enough time – partly because they have not really estimated the value of the debt, meaning what is the risk associated with a different class of debt. Let us say, if you’ve got 100 crores of bonds which are say rated at BB and let us say it is an 8% bond – what are the chances of a default on that? Is it more, or is it less than if you had 8% on let us say an AAA bond. Is it more or is it less? How much more or how much less? This kind of arithmetic and details for it are not there. And that is why we believe that our system has met our needs; there have been no failures of insurance companies unlike in the west – in the west, every year, about half a dozen insurance companies collapse, and therefore it’s a far greater concern for them.

AK : They say technology could be a crucial driver in sort of balancing this risk. Now, what would be needed to business assure these technologies? Would you like to add something on that?

JHN: I don’t think it is much of a technology issue. It’s a data issue, a question of data reliability. The calculation of a risk is no big deal. If you’ve got the statistics, it’s just a fifteen minute job. So it’s not really a technology issue. It’s an issue of whether we have the quality and depth of the data we require, and have we captured all transactions accurately at a point of time. And do it systematically, you know, quarter by quarter, month by month, or even daily for that matter!

To some extent, the Reserve Bank of India does it, and therefore what they have done for different classes of assets held by banks, they have calculated what should be the risk weights associated with different classes. So the IRDA had considered that, and based upon the expertise of the RBI on such matters, they had calculated and sent an exposure draft, saying that we could recalibrate our estimates based upon the risk weights as calculated by the Reserve Bank, and then require companies to make provisions accordingly. The question really is this: Companies are trying to balance between the need for security of an investment portfolio – rather, of an insurance portfolio, at the same time, to minimize the capital required to do so. Capital is expensive. So in countries such as in Europe, what they found as a result of the 2008 issue was it’s always in the shareholders’ interest to keep the capital as low as possible. So they used to keep it very low. They were skating on thin ice. And that is what they are trying to redress. So one of the concerns which Europe and European insurance companies have, is that they believe the Solvency–II requirement might require companies to recapitalize quite substantially, and that poses a problem for shareholders, for the economy at large, and that is the question which they are facing.

AK : Thank you Sir, have a nice day!

No comments:

Post a Comment