This is our third article in our series on Cashflow Driven Investment (CDI). In the first article, we warned that the generally underfunded and increasingly cash flow-negative U.K. defined benefit (DB) pension sector could see its funding status challenged, especially in today’s late-cycle environment. We argued that a CDI approach could be a solution, as a core allocation to income generating/self-liquidating assets could help meet liability payouts with a high degree of certainty. In the second article, we argued that from the standpoint of diversification and the sheer size of the available asset pool, core CDI portfolios require a global investment opportunity set. Given that the objective of CDI is to generate predictable cash flows, the question of how future foreign cash flows can be hedged becomes crucial. In this article, we explore the relative effectiveness of hedging strategies that schemes can employ to seek predictable cash flows in sterling.

Let’s take a simple example of a sterling based cash flow-driven investor who expects to receive¹ $100 in 10 years’ time. With the current market-implied 10-year forward exchange rate of 0.757 pounds per USD, an investor, by entering into a 10 year forward foreign exchange (FX) contract, has the certainty of exchanging $100 against £75.7 in 10 years and therefore perfectly hedging their expected cash flow in sterling.

However, the challenge CDI faces is that long-dated forward FX contracts are at best highly illiquid and therefore the only practical solution is to dynamically replicate the long-dated FX contract with liquid instruments.

Figure 1 illustrates how to decompose the market implied 10-year forward exchange rate into its primary building blocks, namely spot exchange rate, 10-year sterling yields, 10-year USD yields and currency basis.

It follows that investors are faced with three types of risks when looking to hedge foreign bond cash flows into sterling:

1. Spot foreign exchange (FX) risk
2. Interest rate risk differential: In our example, this means fixing the difference between the 10-year USD yield of 1.22% and the 10-year sterling yield of 0.35%.
3. Currency basis risk

If left unhedged, these risks may affect the predictability of the realised cash flows in GBP. The question is how large can the impact be?

In the next section, we explore various hedging strategies and propose a measure of cash flow tracking error that each approach can generate.

• Strategy 1: This is the simplest approach. It consists of rolling liquid short-term FX hedged (typically 1–3 months’ maturity) contracts to immunize the portfolio against variations in the exchange rate. However, this approach leaves both interest rate differential and currency basis risks unhedged.
• Strategy 2: In addition to hedging short-term currency exposure as above, this option hedges the portfolio against changes in USD interest rates swaps (i.e., it hedges USD duration). However, it does not lock in the differential between the USD and GBP yields.
• Strategy 3: This option hedges currency exposure and swaps foreign interest rate risk for sterling duration, thereby locking in the interest rate differential over the entire cash flow horizon; however, this strategy keeps the basis risk unhedged. It is achieved via short-dated FX forwards and a combination of “receive fixed GBP interest rate” and “pay fixed USD interest rate” swaps.

Here is a summary of how each strategy may address each risk:

Using proprietary analytics², we measured for each dynamic strategy³ the deviation between the realised cash flows in sterling and those predicted by a perfect hedging strategy at portfolio inception.⁴  We express this difference between predicted and realised cash flows as an equivalent annual adjustment to the initial yield of the foreign portfolio. Put simply, if the hedging strategy resulted in a cash flow shortage, where realised cash flows were less than the predicted cash flows, it would produce a negative adjustment, and vice-versa. We report the standard deviation of this adjustment in the table below, which can be thought of as a measure of cash flow tracking error.

In our analysis, we assumed the portfolio to be 100% USD-denominated, generating 15 years of equal USD cash flows (each an annual payment worth $100). Figure 3 illustrates the contractual, expected cashflows in USD (blue bars) and the predicted GBP cash flows at inception (green bars).

We tested our three strategies against this scenario and found the following estimates of standard deviation for the annualised yield adjustment.

With an annual standard deviation near 50 basis points, we found that leaving the interest rate differential unhedged (strategies 1 and 2) risked compromising the predictability of the cash flows delivered in GBP, challenging the primary objective of a CDI portfolio, which is to guarantee the future payment of liabilities. In contrast, Strategy 3, which only leaves the currency basis unhedged, offered substantially less volatility than the other two approaches.

In light of these results, and whilst recognizing the case illustrated here is an extreme one as a global CDI portfolio would typically not be 100% denominated in USD, we would strongly recommend that global CDI portfolios are both currency hedged and duration hedged back in to sterling, whilst the remaining currency basis risk can be managed on a tactical basis and form part of the active toolbox of a skilled asset manager.

To find out more about CDI and how PIMCO implements it, please contact Michael Burns – Michael.burns@pimco.com.

ENDNOTES