Two curves and non deliverable interest rate swaps

I have blogged about the importance of two curve discounting in valuation of swaps (here and here), and I have separately blogged (here, here, here and here) about the growing offshore market in rupees and other emerging Asian currencies. But it was an alert reader of the blog who pointed out to me a very interesting connection between the two. The instrument that lies at this intersection is the non deliverable interest rate swap (NDIRS). This is like the non deliverable forward (NDF) market in that it is cash settled in US dollars and operates out of reach of the local regulators. But the underlying product is an interest rate swap rather than a forward contract. The two parties agree to exchange a floating interest rate for a fixed interest rate in rupees or renminbi or other emerging market currency. However, the contract is non deliverable and is therefore cash settled in US dollars without any cash flows in the underlying currency.

The issue is about valuation of the swap and the complexity arises because in India and in many other emerging markets, the OTC derivative market runs without collateralisation. As I explained in my earlier blog post, this means that the valuation depends on the funding cost of the counter parties involved. The problem affects the valuation of the swap even at inception (particularly when the yield curves are quite steep), but the problem is most acute and clearly understood for a swap which has moved into or out of the money some time after inception.

For example, consider a hedge fund that entered into a five year swap agreeing to pay a fixed rate of 5.25% and received floating. After some time, suppose that the swap rate has moved to 5.75%. The hedge fund can now lock in a risk free profit of 0.50% per year for the next five years by entering into another swap in which it receives fixed at today’s rate (5.75%) and pays floating. The net effect of the old swap and the new swap is that the floating legs cancel and the hedge fund simply receives 0.5% fixed for the next five years. The question that arises is what should the hedge fund receive if it wants to unwind the two offsetting swaps and simply pocket the entire profit upfront instead of letting the profit trickle in over five years.

The simple and obvious answer of course is that the hedge fund should receive the present value of the annuity of 0.50% a year. The tricky part is to agree on the discount rate for determining this present value. If the hedge fund goes to a local bank that funds itself largely in the onshore market, the answer would clearly be a discount rate based on the onshore interest rates. On the other hand, if the hedge fund goes to a foreign bank that funds itself largely in the offshore market (borrowing US dollars and swapping into rupees or renminbi), then the discount rate would be based on the all-in (swapped) cost of the offshore borrowing. This latter cost of funds is given by the cross currency swap rates (where a US dollar floating rate is swapped for fixed rate in rupees or renminbi).

In countries with capital account convertibility, there is not too much of a difference between the onshore swap yield curve and the cross currency swap curve because of covered interest parity (CIP). But CIP is clearly not applicable to rupees and renminbi! The cross currency swap rate for currencies like renminbi can actually be negative because of the wall of money wanting to speculate on the appreciation of the currency. Even if things do not get that bad, the gap between the two curves can be several percentage points. Swap valuation using the cross currency swap rate is a form of two curve discounting – the forward rates come from the onshore swap market and the discount rates come from the cross currency swap market.

Smart hedge funds know all about this and choose the banks carefully when unwinding trades. It would clearly like a low discount rate when unwinding a winning trade and a high discount rate when unwinding a losing trade. A bank that watches the hedge fund do this probably feels frustrated, but in fact, the bank is not being cheated at all as the unwind is done at the rate offered by the bank. What the hedge fund is doing is to arbitrage between the onshore and offshore markets with their different interest rates.

All this assumes that the banks are smart and know what their funding costs are. While this may be true of the most sophisticated banks, it is certainly not true for all banks. Some banks may also not be fully clear about the difference between average cost of funds and marginal cost of funds. A foreign bank that funds 90% of its rupee balance sheet in local currency deposits and borrowing may think that its cost of funds is the onshore rate. But if it depends on offshore borrowing for the incremental growth of the balance sheet, it may still be true that its marginal cost of funds (which is all that is relevant for valuation and pricing) is actually the offshore rate. Some banks surely get this wrong.

The lesson in all of this is that the non deliverable market is quite a big mess and there is plenty of scope for supplanting this to a great extent with an onshore cash settled exchange traded currency derivative and interest rate derivative market. Without compromising on capital controls, these exchange traded markets would improve transparency and would move the market (and associated high paying jobs) onshore.

Indian uncollateralised derivative markets

An article by Christopher Whittall in the International Financing Review (IFR) about the difficulties in valuing uncollateralised derivatives is of great relevance to Indian OTC swap markets. Christopher Whittall explains in his article that “Unsecured trades now present a serious valuation headache”. FT Alphaville follows up on this story and highlights the problem: “… pricing even the most basic (uncollateralised) swaps is now very complex. … traders just flat refuse to enter into any detail about how they price uncollateralised derivatives nowadays — hardly a positive thing for a market that is regularly accused of being like a black box. ”.

Over the last few years, the two curve discounting model which discounts cash flows using the OIS curve (see my blog post of last year) has emerged as the market standard for valuing collateralised swaps. This technique is not applicable for uncollateralised swaps, and in fact there is no “market” valuation for these swaps because the value depends on the cost of funds of the two parties via the Credit Value Adjustment (CVA) and Debt Value Adjustment (DVA).

Deus ex Machiatto puts the matter succinctly:

A vanilla derivative is a collateralized one under the standard CSA these days (cash collateral in the same currency, daily MTM, daily margin). Anything else is exotic, because it involves an exotic collateral option.

All this is important for Indian OTC swap markets because the market runs largely without collateralisation. While these swap deals are governed by the standard ISDA (International Swaps and Derivatives Association) documentation, most Indian banks do not sign the Credit Support Annex (CSA) that deals with collateralisation. We have tended not to worry too much about the Indian OTC swap market because it is dominated by plain vanilla interest rate swaps. What Whittal and Deus ex Machiatto are saying is that this view is incorrect. Effectively, these are all exotic derivatives because of the lack of collateralisation. I believe that this is correct and the matter needs urgent regulatory attention.

There are three main ways to set things right:

• Indian OTC markets can move towards collateralised swaps. This is simple as it only requires Indian banks to sign the CSA under standard ISDA documentation.
• We can move one step further towards the post crisis global regulatory consensus and adopt central clearing of the OTC swaps. The Clearing Corporation of India (CCIL) already has the ability to clear OTC derivatives. Of course, moving more risks into CCIL would make it even more systemically important than it already is, and regulators would need to be extra vigilant about this concentrated risk.
• The most radical solution is to move more of the markets towards exchange trading and clearing. There is a great deal of experience with this both in the equity markets and in currency futures and there is the added benefit of price and trade transparency.

I believe that it is necessary to move quickly along one or more of
these alternative paths to mitigate the risks in this market.

The Formula That Killed Wall Street is Alive and Well

The Gaussian Copula which used to be the standard model for valuing CDOs has been described as the The Formula That Killed Wall Street. After the crisis, several alternatives to the Gaussian copula have become popular for CDO valuation.

But there are many other areas where Gaussian copulas still hold sway. Last month, the Basle Committee on Banking Supervision published Operational Risk Supervisory Guidelines for the Advanced Measurement Approaches. The paper notes that the most common method of dealing with dependence in modelling operational risk is by use of copulas; and “Of the banks using Copulas, most (83%) use a Gaussian copula.” In addition about 17% of banks, used a correlation matrix which is even worse than a Gaussian copula.

Faced with this clearly unsatisfactory situation, the BCBS pushes back against this in the mildest possible way:

Assumptions regarding dependence should be conservative given the uncertainties surrounding dependence modelling for operational risk. Consequently, the dependence structures considered should not be limited to those based on Normal or Normal-like (eg T- Student distributions with many degrees of freedom) distributions, as normality may underestimate the amount of dependence between tail events. (para 229)

Not only is the Gaussian copula alive and well, the regulators do not seem to feel any sense of urgency in changing this state of affairs.

The political economy of selling gold reserves

For those of us in India who lived through the economic crisis of 1991, one of the images seared into our memory is that of the plane taking off from Mumbai to London filled with gold to be pledged for an emergency loan. This helped create a broad consensus among politicians and bureaucrats to do whatever it takes to solve the crisis and bring back the gold. Almost two decades later, that episode still influences the thinking of policymakers. I believe that it played some role in the government’s decision last year to buy some gold in the IMF gold auction.

Similarly, one of the enduring images of the Asian crisis of 1997 is that of Korean citizens depositing their gold with the government so that the country could pledge this gold for badly needed loans. The Korean people took ownership of the problem and became determined to overcome the crisis.

I was reminded of all this when I read an article in Time (originally published in Die Welt) pointing out that both Greece and Portugal are sitting on billions of dollars of gold reserves even while they are being bailed out by the EU and the IMF.

It appears to me that when countries receive help before they have exhausted their own resources, the moral hazard is exacerbated. Moreover, commitment to painful reforms is likely to be very weak. During the Asian crisis, the IMF was accused of being too harsh. The reality is that while the IMF medicine was bitter (in a few cases, unnecessarily so), it was successful in creating dynamic economies that could put the crisis behind them.

Responding to the criticism that it was subjected to after the Asian crisis, the IMF seems to have turned from a purveyor of bitter medicine to a dispenser of sugar coated placebos. The EU is even less willing to take any harsh action. This is most unfortunate, and I believe years from now, peripheral Europe would wish that they had a sterner taskmaster.