Market risk is an important risk for all financial institutions, and is often the most important. All firms should have clear strategies and policies on market risk. It is also important to recognise the way in which market risk is linked to other risks.
For example, operational failures can often be highlighted in extreme market con- ditions, so it is important to consider the extent of market risk exposure when de- signing systems to limit operational risk. Market risk is also closely linked to credit risk. Not only does credit risk tend to be higher when markets are subdued, but many derivative-based responses to market risk can expose a firm to counter-party risk. In particular, OTC derivatives expose each counter-party to the risk that the other will fail before the end of the contract whilst owing money. One way to deal with this is collateralisation, which is discussed later in this section.
16.2 Market and Economic Risk 433 16.2.1 Policies, Procedures and Limits
The most fundamental aspect of managing market risk is to have clear policies.
At a high level this can include policies on the overall level of market risk that is acceptable by some measure such as VaR. However, it should also include details of what constitutes an acceptable investment, and what limits there are to investments in particular asset classes, individual securities or with individual counter-parties.
In this way policies, procedures and limits are closely linked to diversification – discussed below – and counter-party risk.
A firm’s policies should also include a statement of who can make various in- vestment decisions, and the financial limits on such decisions. This provides the link between market risk and operational risk.
16.2.2 Diversification
A key way to manage market risk is through diversification. By holding a range of investments, exposure to the poor performance of one is limited. Diversification can be measured by the extent to which a portfolio holds assets in different asset classes, geographic regions and economic sectors, either in absolute terms or rel- ative to benchmarks. Factor analysis can also be used to determine the extent to which particular economic and financial variables influence a portfolio of stocks.
If the exposure to one or more factors is thought to be too great, then this implies that the portfolio should be diversified further.
16.2.3 Investment Strategy
This is arguably the easiest way to manage market risk, although the scope for change and the effect of that change will vary across different types of firm. For banks, market risk is often not the greatest risk faced. However, regulations typ- ically have an impact on the attractiveness of different asset classes, with some seriously discouraged or even prohibited. For insurers, the scope for change is con- trolled by the degree to which the assets held are admissible from a regulatory point of view, or on the capital charges that those assets attract. This can mean that as- sets that are relatively similar from a risk point of view are treated in very different ways from a regulatory point of view. The market risk aspect of the investments is secondary to the admissibility or capital charge aspect for insurance companies.
Market risk is often the key risk for pension schemes, so the investment strategy is a key way of controlling the risk taken. However, it is only one aspect and should be considered in the light of the various other ‘levers’.
Investment strategy is often determined using stochastic asset-liability mod- elling. This typically involves maximising the return, defined by some measure
such as shareholder earnings, subject to some maximum level of risk such as a VaR target. In reality, there may well be a number of risk limits that are applied.
16.2.4 Hedging Against Uncertainty
Rather than changing an investment strategy directly, derivatives may instead be used. One approach is to use derivatives to hedge against uncertainty. This means that both losses and gains are reduced. The easiest way to do this is using a future or a forward. Each of these is an agreement to buy or sell a fixed amount of some asset for a fixed price at some fixed date in the future, the delivery date.
Futures or forwards can be used as an alternative to buying and selling securities if the investment strategy is being changed. They might be used if there is a desire to leave a particular stock selection strategy in place, in terms of the actual invest- ments held, whilst changing the underlying asset allocation. Futures and forwards can also be used to change the asset allocation more quickly and cheaply than can sometimes be achieved by trading the underlying securities.
An important point to note about this type of hedging is that it means that prof- its as well as losses are neutralised. This might not be a problem if this issue is understood by all parties, but even if offset by a large profit in an underlying as- set, a large loss on a derivative contract can be unsettling. This is particularly true if the department carrying out the hedging constitutes a separate cost centre to the department holding the underlying asset. Communication is therefore key in these circumstances. It is also worth noting that even though the aggregate position may be neutral, there may be cash flow implications arising from a large loss on a derivatives contract, and the offsetting move in the underlying asset will not be helpful if that asset is not particularly liquid. Furthermore, if an insurance company has an admissible derivative hedging an inadmissible asset, there could be adverse consequences if the value of the derivative falls.
It is also important to recognise that no matter how good a hedge might be in the- ory, uncertainty over the amount of hedging required can reduce the effectiveness of a hedge. For example, a pension scheme might want to hedge a future sale of assets, but may still be in receipt of contributions that are based on the total payroll and so are uncertain.
Differences between Futures and Forwards
Whilst futures and forwards have similar underlying properties, they differ in some important ways. The most fundamental is that futures are traded on exchanges whilst forwards are OTC contracts. Anyone wishing to trade a future must be a member of an exchange or must trade through a broker who is a member. Each futures trade involves matching a party who wishes to take a long position in a
16.2 Market and Economic Risk 435 future with one who wishes to take a short position, since each future is a contract.
However, even though each trade will match these two parties, the parties do not contract with each other. Instead, all parties contract directly with the exchange.
Forwards, on the other hand, are simply OTC agreements directly between the two parties wishing to trade. The details of each contract are typically set out in an ISDA (International Swaps and Derivatives Association) agreement. This is a very detailed document outlining all aspects of how the contract works.
As OTC contracts, forwards are very flexible and can be provided on virtually any underlying asset with any delivery date. However, the bank providing the for- ward will itself want to mitigate this risk, either through other positions held or with other banks, and the more unusual a forward is the more difficult this will be.
More importantly, the more difficult it is to pass on the risk in the forward, the more risk capital a bank will need to write the forward. Since this cost is passed on to the investor, there is a real cost to pay for demanding an unusual forward contract.
Exchange-traded contracts such as futures have virtually no flexibility – they are highly standardised in terms of the nature of the underlying asset and the delivery date. However, this level of standardisation means that exchange-traded contracts tend to be very liquid, meaning that large transactions can be effected very quickly with a minimal impact on the price of the contract.
Counter-Party Risk
The nature of exchange-traded and OTC contracts has an impact on the credit risk faced by the various counter-parties. Looking first at exchange-traded contracts, counter-party risk is reduced by the pooling of contracts – since each party has a contract directly with the exchange, the failure of a single counter-party does not directly affect the payment of any single futures contract. However, since this means that the exchange is underwriting all contracts, the exchange needs to protect itself from the failure of any of its counter-parties – in other words, those holding futures contracts.
Exchanges protect themselves through the use of margins. These are deposits that members of an exchange post with the exchange to ensure that if a member becomes insolvent, there are assets available to cover any losses they have made on their contracts.
There are several types of margin that might be required, the most common being:
• initial margin;
• maintenance margin; and
• variation margin.
The initial margin is the value of assets transferred to a margin account once a
contract is opened. This will be some proportion of the contract size, with the proportion depending on the volatility of the contract. At the end of each day – and sometimes during the day – the cost to the member of closing out a position at the current price of the future is calculated by the exchange. A futures contract is closed by taking an opposite position in the same contract.
If the cost of closing the position would be greater than the initial price of the future, and the position in the future is a long position, then the difference is de- ducted from the margin account; if it is lower, then the difference is added to the margin account. This process is known as marking to market.
If the margin account drops below a specified level – the maintenance margin – then the member is required to transfer assets to the margin account to top it back up to the level of the initial margin. This amount is known as the variation margin.
Margins can be reduced if members hold diversifying positions in similar fu- tures – that is, if they hold spread rather than naked positions. At the extreme, this can involve margins being calculated taking into account the individual con- struction of each member’s portfolio with the exchange. This is the case with the Standard Portfolio Analysis of Risk (SPAN) developed by the Chicago Mercantile Exchange.
Each exchange will specify what assets can be counted as collateral. They will also specify the extent to which each asset counts. For example, high-quality gov- ernment bonds might be counted at 90% of their face value whilst shares might only be counted at 50% of their face value.
With the absence of pooling, counter-parties to OTC derivatives such as for- wards face a higher degree of counter-party risk. This is often dealt with using collateralisation. Collateralisation involves the transfer in response to the marking to market of a contract in a similar way to margin requirements. The obligations of both counter-parties in relation to collateral are outlined in the credit support annex (CSA) of the ISDA agreement. In particular, whilst the ISDA agreement covers all aspects of the structure of the derivative and the calculation of its price, the CSA covers issues such as the types of security that can be used as collateral and when the required amount of collateral is calculated. It will also specify the minimum transfer amount – the level below which no transfer of collateral will be needed.
This is to avoid counter-parties making very small transfers of assets when marking to market reveals that only a small change in the collateral required is needed.
Not all OTC contracts will involve collateralisation; however, collateralisation can reduce the cost of a transaction for a counter-party whose risk of default is regarded as significant.
Even if adequate collateral has been posted, the failure of a counter-party can leave a firm exposed to a risk that it had hoped to deal with. If the failure has occurred at a time of more general difficulties in the market, then putting a replace-
16.2 Market and Economic Risk 437 ment derivative contract in place might take some time, leaving the firm exposed to risk for longer than it would prefer. The contract may also be more expensive to replace than it was to put in place originally. These factors must be borne in mind when considering how to deal with market risk.
Pricing Futures and Forwards
If costs are ignored and the asset on which the future is based pays no income, then the price of a future or forward has a simple relationship to the spot price of the underlying asset. In particular, the price at time 0 of a future or forward with a delivery timeT,F0, is related to the spot price at time 0,X0, and the continuously- compounded risk-free rate of interestr∗as follows:
F0=X0er∗T. (16.1)
In other words, the price of the future – which represents the price at which an investor is agreeing at time 0 to buy or sell an asset at timeT – is simply equal to the current spot price rolled up at the risk-free rate of interest. The rationale for this formula can best be seen by considering two equivalent ways of owning an asset at timeT. The first is simply to pay the spot price for the asset,X0, at time 0; the sec- ond is to enter into a futures contract at time 0 to payF0for the asset at timeT, and to invest sufficient assets in an account paying a risk-free rate of interest to accu- mulate toF0at timeT. This would require an investment ofF0e−r∗T. Since the two transactions must have the same price – otherwise an arbitrage opportunity would exist – this means that F0e−r∗T =X0 which, after rearrangement, is equivalent to Equation 16.1.
In practice, there are often complications. In particular there may be:
• a fixed amount of income available from the underlying asset;
• a fixed rate of income available from the underlying asset;
• a fixed amount of benefit associated with the underlying asset;
• a fixed rate of benefit associated with the underlying asset;
• a fixed amount of cost associated with the underlying asset;
• a fixed rate of cost associated with the underlying asset; or
• differential rates of interest for inter-currency contracts.
If a fixed amount of income is payable on the underlying asset, then this is nor- mally foregone if the investment position is replicated by a future or forward – such a derivative commits the holder to trade in the asset, but does not result in a transfer of the asset’s income in the period before transfer. To allow for the lack of an amount of income with a present value at time 0 ofD, this amount must be deducted from the spot price of the asset. This means that the price of the forward
becomes:
F0= (X0−D)er∗T. (16.2) If income is instead received at some fixed rate,rD, then this rate is instead deducted from the rate at which assets would need to accumulate:
F0=e(r∗−rD)T. (16.3) As well as income, holding an asset can provide other benefits. For example, hold- ing a physical asset rather than obtaining the position synthetically might result in a reduction in capital requirements. If such an effect is market-wide, then it could have an effect on the price of a future and can be reflected in Equations 16.2 and 16.3, with the benefit being converted to a fixed amount,D, or a rate rD. The rate of benefit is known as the convenience yield.
Assets can also generate explicit costs, which are not borne if exposure is gained through a futures contract. For commodities, this can be storage costs, but it might also be a cost of financing. Such costs can be regarded as negative income in Equa- tions 16.2 and 16.3.
As a result, the price of a future can be above or below the current spot price. It can also be above or below theexpected future spot price. If the price of a future is lower than the expected future spot price, then the situation is described asnormal backwardation. This occurs when any income produced by an underlying asset together with its convenience yield exceed any storage or financing costs. However, it also occurs if the main reason for the existence of the market is for producers of a commodity to hedge against future falls in the commodity’s price – the volume of demand for short positions in futures drives the price of the future down.
The opposite situation occurs if the market is driven by a desire to gain expo- sure to an underlying asset synthetically, so there is a high volume of demand for long positions in futures. This can happen if the reason for the creation of a mar- ket was to allow users of particular commodities to hedge their input costs, or if prices are driven by investors trying to gain exposure to particular commodities.
This also means that the effect can be exacerbated if storage costs are particularly high. In this case, the market is said to be incontango. Both markets are shown in Figure 16.1, withXtT denoting the expectation at timetof the spot price at timeT. These terms should not be confused with normal and inverted market. A normal market is one where, on a particular day, the futures prices increase with expiry date of the future. This is what might be expected when there is no predictable sea- sonality in the availability of the underlying asset. However, for some commodities there might be an expectation of increased availability of the underlying asset at some future date. This could be expected to lead to a fall in the spot priceat that time, which would be reflected in a lower price for futures of that term.
16.2 Market and Economic Risk 439 Contango
Price
Time
0 T
Ft
XtT
Normal Backwardation
Price
Time
0 T
XtT Ft
Price
Time
0 T
Figure 16.1 Markets in Contango and Normal Backwardation
Basis Risk in Futures
The point of a forward is that it can be used to hedge exactly the risk faced, with the size of the contract being equal to the size of the risk. However, the fact that futures contracts are standardised means that they might not provide an exact hedge. In particular:
• the futures position might need to be closed before the expiry date of the future;
• the future may expire before the planned date of the asset’s sale or purchase, requiring that the future be rolled over into another position;
• the date of sale or purchase for the asset might be uncertain;
• the asset on which the future is based might not be the same as the asset being hedged; or
• items excluded from the future such as dividend income from the underlying asset or costs associated with investment in this asset might not be known accu- rately in advance.
All of these issues can give rise to basis risk, which is defined as uncertainty in the basis at the point at which the futures position is closed. The basis at timet,Bt, is the difference between the spot price of the asset,Xt, and the futures price,Ft:
Bt=Xt−Ft. (16.4)
As can be deduced from the earlier comments on discounting, storage costs, conve- nience yield and so on, the basis can be positive or negative. It can also be defined asFt−Xt, particularly in the context of financial futures.
The basis at the time a futures contract is effected is known, since both the price of the future and the spot price of the asset are known. Furthermore, if a hedge is required until the exact date of expiry, T, the asset being hedged is exactly the same as that underlying the future and there are no uncertain cash flows in the