Companies Want Out Of Pension Business
Anthony Gould is managing director of J.P. Morgan Asset Management. He was in Toronto recently and took some time to meet with Joe Hornyak, executive editor of Benefits and Pensions Monitor to discuss liability driven investment (LDI) in 2013,
Benefits and Pensions Monitor: Where are we in the world of LDI these days?
Anthony Gould: Most plans understand finally that this, the end game they want to get to, is substantially de-risking their portfolio and removing, in particular, a lot of the equity risk that they have in portfolios.
Because of the regulatory and accounting changes over the last few years, as well as the volatility that we have seen in capital markets, a lot of companies that we speak to just want to get out of the business of running pensions. Pensions have been much larger headaches then they ever imagined. Companies that for many years have not made contributions to their plans have been forced to make these contributions.
Because of accounting changes, and I am talking broadly because we have seen this in many countries, in Canada, in the U.S., in the UK, in the Netherlands, we have seen very similar changes which has increased the transparency in financial statements. Analysts and investors can all see very clearly the impact that changes have in pension solvency funded status and the impact of those changes on the net worth of organizations on volatility of earnings. As a result, a lot of CFOs and treasurers are saying we want to reduce the impact that pensions are having on our financial statements, on our cash flows.
The way to do that is to really de-risk, which means reducing risky asset exposures and increasing the allocation to long duration fixed income.
BPM: Has the upturn in equity markets over the recent months put plans in a position to de-risk and are we starting to see plans move that way? Or are they still facing that barrier of low interest rates?
AG: We have recently seen a particular interest just over the last few weeks. Obviously, the uptake that we have had in yields just in May will probably accelerate that a little bit. I think the reason why we are seeing this despite – the still relatively low level of rates – is because quite a few plan sponsors have put glide paths in place. What glide paths tend to keep to is funded status. What a glide path can say is my funded status is maybe today 80 per cent, but when it gets to 90 per cent, I am going to de-risk the plan a little bit so I am going to increase my allocation to fixed income. It makes sense to key those glide paths to funded status because that affects all of the metrics that CFOs and treasurers care about.
BPM: Are these funds that made a commitment to LDI two, three, four years ago had glide paths in place since then?
AG: Some have done so more recently. Even over the last six to 12 months, we have had conversations with plans that say ‘well, we are thinking about glide paths’ or maybe ‘we put a glide path in place and we want to talk about actually how to implement that.’ A lot of glide paths we have seen tend to define only very loosely in terms of this is the target allocation to equity. This is the target allocation to long duration fixed income.
BPM: How long does it take to put together a reasonable glide path program?
AG: In our experience, the heavy lifting, the quantitative work, is not that complex. What often takes more time is getting organizations, getting investment committees and investment staff, comfortable with that because of the way a lot of organizations are set up. There may be an investment committee with trustees of the pension plan that typically have had the final say on asset allocation. So for them to buy into a glide path means they feel like they are delegating away or discharging their responsibility for asset allocation to some kind of automatic process. Just politically, in some organizations, it may be tricky for them to feel comfortable doing that.
BPM: Are we in the media too focused on LDI?
AG: No, I think this is the right focus because LDI obviously involves generally dialing down the exposure to risky assets and dialing up the exposure to fixed income. In terms of how that is being done, I would say that generally depends on where the funded status is. So plans that are still substantially underfunded need to have a significant allocation to risk assets. They may feel, given the pain that they have experienced, that they want to reduce their duration mismatch. So those kinds of plans have looked at, and we have clients that have done this, interest rate overlays and derivatives as ways to reduce that duration gap.
BPM: We see that some of the larger plans – Teachers’, OMERS, HOOPP – have moved to asset liability matching strategies using the private equity markets. Is that an option for LDI or for private sector plans that want to de-risk their funds using LDI?
AG: It makes sense for public sector plans. We see this in Canada, in Australia, and, to a degree, in the U.S. We also see it in the UK because many public sector plans in many countries have COLAs (cost of living adjustments), so effectively their liability interest rate sensitivity is to real rates rather than to nominal rates. I would say that in the first instance.
The second thing is in the public sector, of course, in most of these countries, plans are still open and will remain open probably for the foreseeable future. Public sector unions probably have enough sway so that those plans will remain open. Contrast that with private sector plans where in some countries there is no firm commitment to provide inflation sensitivity, inflation linking, to the retirement benefit after retirement in some countries, not in all countries. Therefore, the interest sensitivity of liabilities is nominal rather than real. In that case, real estate and infrastructure projects are just less suitable because they are providing real rate exposure rather than nominal rate exposure.
The second aspect is we are seeing more and more private sector plans close or freeze. For those closed or frozen plans, very long duration projects, which have little to no liquidity, such as infrastructure, in particular, are probably of less interest.
BPM: So the approach to LDI changes if your plan is closed or frozen?
AG: Yes, it would, but it would depend on the kind of freeze. In the hardest freeze, there is no service cost, no accrued liability, it is literally fixed. In that case, it is just like a bond obligation, cash flows. There is some actuarial uncertainty, but aside from that, the cash flows are pretty much known. Therefore, the right match for that kind of a portfolio is going to be just 100 per cent fixed income.
For a plan where there is still some service cost of accrual of benefit, then you think about what your required rate of return for that plan, which, let us assume, is fully funded. If you do not want to continue to make contributions to that plan, then you need a return that is in excess of just a fixed income or the rate that the liability is accruing without the service cost. So you need a bond plus kind of a return. There, you may have significantly less than 100 per cent of your assets in fixed income because you want your assets to help to fund the cost of those accruing benefits.
The less frozen the plan is, generally, the less it is going to have LDI kind of assets or strategies.
BPM: One of the unique characteristics of Canada is that its fixed income investors are 99.9 per cent invested in Canada. They are not going outside the country. Is that true in other parts of the world or is it unique to Canada?
AG: You do see home country bias in a lot of countries. Every market is unique. There is a tendency to generalize and say what are pension investors doing globally. The reality is pension investors in each country have their own regulatory framework. They have their own cultural and legal history, have their own accounting standards to deal with.
In Canada, of course, going back a long time now, there was this ceiling on the foreign investments in a portfolio.
Back when I was a bond manager in Canada 20 years ago, the way that Canada pensions dealt with that issue was by allocating up to the ceiling in non-Canadian equities. They did not want to use that with non-Canadian fixed income. So some Canadian managers would get around that by using some derivatives to get exposure to U.S. bonds.
I wonder whether that kind of history has coloured the approach that Canadian pensions take with respect to looking at non-Canadian fixed income because when I was back in the UK in the late ’80s, even then, UK plans had as much or more in non-UK fixed income as they had in sterling fixed income. In Australia, the market grew so fast with the mandated superannuation plans that the in-flows of foreign fixed income were very large compared to new Australian issuance. So maybe in each individual country there is a reason why they have allocated more to non-domestic fixed income.
BPM: Where are pension funds going to find the fixed income products to de-risk their portfolios?
AG: In the U.S., since the credit crisis, long corporate bonds have been added. So coming out of the credit crisis, credit spreads were at the widest levels since the 1930s. From a value perspective, there is a good reason to look at credit, corporate bonds.
The second reason is the U.S., like virtually everywhere else, from an accounting perspective, requires that liabilities be valued using a high-quality corporate curve. Actually in the U.S., pension regulations also require the use of corporate curve. In Canada, and in some other countries, there is a requirement to use a government curve for solvency purposes. From a liability matching perspective, as well as from a valuation perspective, corporates are very attractive and that has continued as yields have fallen. Plans are saying we would rather go with a portfolio of long corporates, which yield four to 4 1/2 rather than a long treasury portfolio yielding 2 3/4 or three. A lot of interest in the U.S. has been in the corporate ones.
In the UK, historically, a lot of LDI has been done with futures, interest rates, swaps, even total return swaps, some of them through government bonds. The problem with the sterling corporate market is even through there is more than $200 billion outstanding in long sterling corporates, the market is not very liquid. That tends to lead to plans, and actually we have seen a lot of insurance companies, allocating increasing amounts to, especially, U.S. dollar credit because that is where the liquidity is.
BPM: That raises an interesting conundrum if the trend we are seeing to de-risk by buying annuities for either your complete plan or your retiree portion continues, that could create even more competition for these corporate products, would it not?
AG: Absolutely, with the annuitization, obviously the insurance company has to go out and hedge that and the hedge for them is going to be, generally, to buy an investment grade corporate portfolio.
BPM: What will that do for the prices of corporates?
AG: We have seen that credit spreads in the are U.S. very well supported. That is the case in the UK as well and we think that will continue just on an ongoing basis as there is a very strong demand for high quality long corporate paper.
You think about the demand supply imbalance, there is about $1.1 trillion in long U.S. investment grade corporates, which is roughly only about half the size of the U.S. corporate defined benefit market. So if all corporate plans in the U.S. decided that they wanted to embrace LDI, there is a substantial shortage and that does not even take into account what we are seeing, which is liability matchers globally looking to U.S. credit as a hedge. We are seeing non-Japan Asian insurance companies allocate to long credit, long U.S. dollar credit. We are seeing interest from pension plans in the UK and in Canada allocate to global and U.S. long corporates.
BPM: So DB plans anywhere that are looking to use LDI had better start doing it or there may no longer be an affordable option for them?
AG: The wonderful thing about markets is, of course, it is where demand and supply are brought together. So presumably there is price at which they will be able to find that exposure. It is certainly true that the last plan that de-risks is presumably going to pay a significantly higher price than the plan that de-risks today in terms of credit spreads. Where risk free bonds trade is going to be is largely a function of where the economy is, what is going on with short rates, and whether the central banks decide that they are going to start backing away from quantitative easing and slow down or stop the purchases of long-term bonds. Obviously, what is going to be very important is setting risk free rates, but where credit spread are relative to risk free rates is certainly greatly influenced by ongoing demand from pensions and liability matchers.