3.1. Products

With-profits and fixed annuities were the most common products offered by respondents. There was also a broad range of other products (Figure 3).

Figure 3

Products offered by respondents; DB pensions, defined benefit pensions

6.1. Hedging Policy Documentation

We first asked whether companies had a formal hedging policy document in place. Less than half had a comprehensive document covering all hedges within the company. For those that did not most had some form of documentation – either covering most hedges, or a strategy outlined in another formal document. There was only one respondent without any hedging policy documentation, and one where this was work-in-progress (Figure 9).

Figure 9

Hedging policy document existence

However, for many hedging policy documents it is not clear from our data how comprehensive are these documents. The “risk factors to be hedged” is the only feature common to all hedging policy documents amongst respondents. Many items that one might expect to see as best practice within a hedging policy document did not feature in some responses.

An interesting point to note is that the level of detail in the hedging policy document does not appear to be correlated with the size of company. We might have expected larger companies to have more explicit hedging policy documentation. However, this is not supported by the survey results.

The variable annuity providers had more explicit policies, in keeping with the more sophisticated techniques used (Figure 10).

Figure 10

Hedging policy document scope

6.2. Hedging Process Management

We looked at the function that undertakes each step of the hedging process. As you might expect, hedging recommendations are typically made at a less senior level than hedging approval decisions. In turn, these approval decisions are typically made at a less senior level than approval of changes to hedging policy.

There appears to be a range of delegation within insurance companies. In one outlier, the Board is involved with recommending hedging decisions. In some small limited cases, approving hedging decisions do not reach any formal committee.

For the most part, hedge decision approval is undertaken at committee level. For half of respondents management committees also approve changes to hedging policy, without any requirement for this to go before the Board.

As we might expect, it is the ALM functions and ALM committees where hedging activity most often resides within a company. The risk functions and risk committees appear to have much less involvement, despite hedging objectives generally being about the management of market risk. We are conscious, however, that the ALM and risk committees may be “roses by other any name” with different naming conventions across companies (Figure 11).

Figure 11

Seniority and function of hedging process; ALM, asset liability management; WP, with profits

6.3. Hedging Restrictions

No company gave 100% discretion to the hedging manager for any particular risk factor. Some companies do give discretion to hedging managers, subject to the limits of the hedging policy document – incrementally more so for unrewarded risks of rates, currency and inflation (Figure 12).

Figure 12

Hedging discretion (as % of those with material risk)

A number of respondents had restrictions on the type of instrument that could be used for a variety of reasons including regulatory restrictions, modelling limitations and lack of asset management capabilities. For quite a wide range of instruments there were a small number of companies in each case that were restricted. We could not see any discernible pattern with there being restrictions for each type of asset exposure – currency, equity, credit and interest rates – and for derivatives both with and without optionality (Figure 13).

Figure 13

Instrument restrictions

6.4. Drivers of Changes to Hedging Policy

We asked respondents what the strongest drivers of change to hedging strategy are and the results are shown in Figure 14. First, it should be borne in mind that the survey took place in late 2015 (immediately prior to Solvency II implementation) when many companies would have been in a state of transition.

Figure 14

Drivers of changes to hedging policy; FX, foreign exchange

As might be expected, the regulatory environment was the greatest push factor. It is interesting to note that regular reviews are also a significant driver, suggesting that there is not necessarily a “keep in the drawer” solution.

The following graph provides a heat map presentation of results. For each company response we assigned the following weights:

• ∙ 100%=“likely” to “explicit”

• ∙ 50%=“maybe”

• ∙ 0%=“no”

The table then gives an average weighting for each reason, both as a total, and also split by risk factor.

6.5. Variable Annuity Firms

6.6. Acceptance of Risk

We asked companies why they would be willing to accept risk with the following responses (Figure 15):

Figure 15

Reasons for accepting risk

Companies that answered that the risks were not well understood were also those that required Board approval for hedges and had less explicit hedging policies. Only half of these companies answering also answered that risks are not significant enough suggesting that there is scope for greater delegation of hedging.

7.1. Equity Risk

For those respondents that hedge equity risk there is roughly a 50:50 split between those that choose to only delta hedge and those that additionally hedge “second-order” Greeks.

For second-order hedging, it is typically a combination of both vega and gamma that is used. It may be that only one of these is the defined hedging measure, with the other being hedged implicitly as a result. There is a slight bias towards vega – that is exposure to movements in volatility – as the preferred second-order hedging measure.

These results are consistent also with typical practice in the variable annuity industry, which are some of the most active/dynamic hedgers of equity risk.

Those that regarded equity risk as material were equally likely to be delta only hedgers as those that hedge the second-order impacts. Those that regarded equity risk as of medium materiality were more likely to hedge gamma/vega. Although not statistically significant this can suggest that the size of risk is not necessarily correlated with the potential change in the size of risk (Figure 16).

Figure 16

Equity “Greeks”

Equity futures are the most popular choice of instrument, with equity options and notional shorting (by with-profit companies) also being popular (Figure 17).

Figure 17

Equity instruments; ETFs, exchange-traded funds

Of the six with-profits companies that responded, four used notional shorting with three of these in conjunction with futures.

7.2. Credit Risk

The main findings for credit risk are:

• ∙ Surveyed companies have a material exposure to credit risk.

• ∙ Risk is mostly not hedged or is managed within with profit funds via notional shorting.

• ∙ Hedge is defined as tactical or tail.

• ∙ Target of the hedging are credit spread movements and defaults.

• ∙ Single name and index credit default swaps (CDS) are used.

In total, 11 companies said they had material credit risk with eight providing details about their hedging strategy.

Only four of these companies said they did some form of credit risk hedging with one additional company using notional shorting. The more complex derivatives (tranche protection and credit option strategies) are used by companies who are hedging tail risk or extreme spread movements (Figures 18 and 19).

Figure 18

Credit risk hedging (number of respondents)

Figure 19

Credit instruments; CDS, credit default swaps

7.3. Interest Rate Risk

The main findings for interest risk are:

• ∙ All companies are exposed to interest rate risk, most of them considering it a material risk and having medium or high exposure.

• ∙ All companies hedge interest rate risk in some measure.

• ∙ Most companies have a dynamic hedging programme in place. Companies selling variable annuities are all in this group while companies selling with-profits have a more static approach.

• ∙ Rebalancing happens daily or even intraday for almost half of the companies.

• ∙ All companies hedge duration and some also hedge convexity and volatility. Cash flow hedging is also used by both annuity and with-profit companies.

• ∙ Bonds and swaps are the most popular instruments used to mitigate the risk. Swaptions are also widely used.

Most companies have a dynamic hedging programme and these are mostly companies with a high exposure to the risk (Figure 20).

Figure 20

Hedging strategy versus materiality of interest rate (IR) risk

The frequency of the rebalancing of hedges was not uniform with most companies hedging at least monthly. As expected, one driver is the type of product with companies with a variable annuity business hedging more frequently and companies with a with-profits business less frequently. Those hedging semi-statically were the companies with a with profits business (Figure 21).

Figure 21

Product type versus hedge monitoring frequency; WP, with profit

The hedging target for all companies includes duration. This is monitored mainly using bucketed sensitivities (PV01 term structure). Convexity is hedged by only three of the 11 companies. Two with-profit firms also hedge interest rate volatility through the use of swaptions.

Cash flow matching is only used by three companies. Interestingly, at the time of the survey not all annuity companies looked at cash flow matching. This was unexpected given the requirements for MA portfolios (Figure 22).

Figure 22

Methods of hedging interest rate risk

Bonds, swaps and swaptions are the most popular instruments used to hedge interest rate exposure with most companies using a combination of these (Figure 23).

Figure 23

Interest rate hedging instruments used; IRS, interest rate swaps; TRS, total return swaps

7.4. Currency Risk

The main findings for currency risk are:

• ∙ Exposure to currency risk is relatively low for most firms.

• ∙ Hedging is mostly dynamic with daily or weekly rebalancing for most firms.

• ∙ All companies except one use FX forwards to hedge their exposure. Cross-currency swaps and FX futures are also used.

• ∙ Companies do not consider cross-currency basis to be an issue. They have small exposure to the risk or have other mechanisms to absorb it.

In total, 10 out of 14 companies are exposed to currency risk and eight provided details about their hedging strategy. Most of them consider their exposure to be medium or low, with only one company selling annuity products with high exposure to the risk. Hedging is mostly dynamic with five companies hedging at least weekly. Two companies have a less dynamic approach and hedge only monthly or even less frequently.

Rolling FX forwards (seven responses out of eight) are the most popular instrument for hedging credit risk with futures (three) and cross-currency used to a lesser degree (four). Even when cross-currency swaps are used this seems to be in conjunction with FX forwards. It is interesting that in the absence of any regulatory push the preferred method of hedging is via a rolling programme (Figure 25).

Figure 25

Methods of hedging currency risk

Most companies do not consider currency basis risk in their hedging programme. Of the three companies who consider it, two have a mechanism that would absorb the volatility generated by the movement in the basis (liability discount rate or hedge accounting). The other does not consider it to be material (Figure 26).

Figure 26

How companies manage cross-currency basis risk

7.5. Inflation Risk

The main findings for inflation risk are:

• ∙ Exposure to inflation is low for most surveyed companies.

• ∙ Hedging is done less frequently than monthly by most firms.

• ∙ Exposure is measured mostly with term structure sensitivities (IE01 by term).

• ∙ Bonds and swaps are used to reduce the risk.

• ∙ Inflation optionality is delta hedged in all cases apart from one company who is using limited price indexation (LPI) swaps.

Most surveyed companies have some exposure to inflation risk (seven out of 11) but do not hedge as closely as for interest rate or FX risk. Hedging is less dynamic and less frequent than interest rate risk with only three companies (out of seven who hedge the risk) hedging dynamically and only two companies hedging at least weekly. The most common type of product that is inflation linked is annuities.

Exposure to inflation is usually measure in terms of IE01 per bucket. One company relies solely on cash flow matching. One company also uses a top-down approach in conjunction with IE01 (Figure 27).

Figure 27

Methods of hedging inflation risk

Participants mainly use government bonds and swaps to hedge the risk (Figure 28).

Figure 28

Inflation hedging instruments used; IL, inflation-linked

Four respondents have material inflation linked options. Of these, only one hedges the risk using LPI swaps. The other three companies delta hedge the risk with RPI linked instruments. None of them uses inflation options.

We see no respondents explicitly using inflation options to achieve cap/collar strategies (e.g. to directly match RPI(0,5) liabilities). This confirms what we might expect given the illiquidity/cost of this market (Figure 29).

Figure 29

How are inflation options hedged? LPI, limited price indexation; RPI, retail price index

8.1. Matching Adjustment

Six companies have one or more MA portfolios. We have investigated their approach to hedging of risks within the MA and how they approached the management of the cash flow matching requirement (Figure 31).

Figure 31

Do you hold a Matching Adjustment portfolio?

There was no consensus on how frequently the MA hedges will be rebalanced. The approaches range from as frequent as weekly, to quarterly, or even taking an event-driven approach. We would expect that MA hedges would tend to be rebalanced less frequently than for other portfolios due to the overarching “buy and hold” management ethos. An event-driven approach might be expected to include the addition of new business to the MA fund or relevant matching statistics reaching a certain threshold (Figure 32).

Figure 32

How frequently do you intend to rebalance the hedges within the Matching Adjustment portfolio?

PV01 sensitivity is the most common approach to measure the cash flow mismatch in the MA portfolio. As we can see from the table the measures are used in combination by four companies (two preferred not to disclose the metric used). Surprisingly two companies do not use the two metrics with a hard PRA limit (Tests 1 and 2), where compliance is specifically required from the PRA. However, we interpret this to mean that these metrics are not primary drivers of decision making, rather than something that is not monitored and adhered too. The PRA Tests are described in more detail in the Appendix (Figure 33).

Figure 33

What metric are you using to measure the extent of any mismatch and hence the need to rebalance? PRA, prudential regulation authority

No respondents stated that they would be changing their inflation hedging approach due to the MA. Four large respondents have said they already hedge it sufficiently, one medium-sized respondent does not consider it a material risk.

Three respondents hedge within each MA portfolio separately, one hedges multiple MA portfolios together in aggregate. Similarly, three respondents hold derivatives within each MA portfolio separately, one holds them for multiple MA portfolios together in aggregate (Figure 34).

Figure 34

Where do you hold the derivatives to hedge the risks of the Matching Adjustment (MA) portfolio? FX, foreign exchange; IR, interest rates

Two respondents hold collateral for derivatives in each MA portfolio separately, one holds collateral for multiple MA portfolios together in aggregate (Figure 35).

Figure 35

How is collateral managed for derivatives in the Matching Adjustment (MA) portfolio?

Three out of four respondents who hedge FX risk use cross-currency hedging at an individual bond level in their MA fund. Only one company has taken a different approach and uses a cash flow hedging at portfolio level. This might imply that a more frequent rebalancing exercise might be necessary whenever a bond in foreign currency is traded.

It is interesting to compare to Figure 25. When considering currency hedging for wider insurance business, outside of the specific MA fund requirements, rolling FX forward is the most popular currency hedging method. The absence of any rolling FX forward approach from MA fund hedging, is a direct consequence of guidance from the PRA in their letter of 28 March 2015Footnote ¹ . In their view, the usage of FX forwards to hedge non-sterling bond exposures does not provide a fixed cash flow (as defined by Article 77b(1)(h) of the Solvency II Directive), and consequently they do not envisage circumstances in which these assets would meet the eligibility criteria for the MA (Figure 36).

Figure 36

What is your view on cross-currency hedging under the Matching Adjustment? FX, foreign exchange