I was surprised to see today that Norcors’s pension trustees calculated fund liabilities as at 1/4/2018 to be virtually the same the IAS 19 accounting liabilities on 31/3/2018, resulting in a very similar deficits on both basis. This is highly unusual, for example I reported yesterday that MPAC’s trustees calculate liabilities about 17% higher than IAS 19, resulting in a significant deficit and recovery payments vs a significant accounting surplus.
Ultimately the only explanation is that the differences between assumptions used by the accountants when calculating IAS liabilities and those used by the trustees are much smaller than with other companies / funds. By why?
I have double checked various sources and confirmed that my understanding that the differences are usually mostly because IAS 19 dictates the use of high quality corporate bond rates to discount the liabilities back to the present, whereas trustees typically use gilt rates. However, investigating further, it seems that this use of the gilt rate by trustees stems mostly from the fact most pension funds are actively being switched into index-linked supposedly in order to de-risk them after equity returns have disappointed from around 2000.
Here are three pension funds and their exposure over time:
|2012||21%EQ, 19%CB, 19%IL, 11%P, 21%AA, 9%C||21%PE, 39%AR, 27%CB, 0%P, 0%IL, 14%C|
|2013||44%EQ, 11%IL, 4.3%CB 10%P 18%AR 12%AA 1%O||27%EQ, 19%CB, 23%LD, 10%P, 20%AA, 1%C||21%PE, 39%AR, 39%CB, 0%P, 0%IL , 1%C|
|2014||32%EQ, 11%IL, 13%CB, 10%P, 14%AR 14%AA 2%O||20%PE, 40%AR, 40%CB, 0%P, 0%IL , 1%C|
|2015||20%PE, 40%AR, 40%CP, 0%P, 0%IL, 1%C|
|2016||25%PE, 23%AR, 40%CP, 5%P, 5%IL, 1%C|
|2017||20%EQ, 25%IL, 15%CB, 10%P, 30%AR, 0%O||22%PE, 27%AR, 39%CP, 5%P, 6%IL, 1%C|
|2018||18%EQ, 25%IL, 15%CB, 11%P, 30%AR, 1%O||18%EQ, 4%PE, 10%P, 3%AR, 7%J, 3%I, 25%IL, 20%CB, 10%CO||21%PR, 27%AR, 40%CP, 5%P, 6%IL, 1%C|
|2019||TBA||18%EQ, 3%PE, 9%P, 2%AR, 7%J, 3%I, 25%IL, 27%CB, 6%CO||TBA|
Key: PE = Private Equity, EQ = Equity, AR = Absolute Return, CB = Corporate Bonds (investment grade), J = Junk Bonds, P = Property, IL = Index linked / liability driven, AA = Alternative, CO = Cash, non-index linked gilts and other.
In the examples below I’m going to count “equities” as PE + EQ + 50% AR. What is described as Absolute Return I have seen vary between equities with hedges to infrastructure to junk bonds.
So in MPAC’s case they have gone from 53% equities to 33%, with increases in index linked and corporate bonds, large increases in corporate bonds and especially index linked to 25%.
BT have been more stable (or perhaps the major shift occurred earlier), but some de-risking trend is clear with them now holding 19% equities and again 25% index linked.
Norcros is, as suspected, is a real outlier with total likely equities only reducing from 40% to 34% and with just 6% now in index linked.
The high proportion of equities and low proportion of index linked would explain why the trustees believe their returns will be higher than other schemes. If they plan to continue this asset mix you can see why they might be willing to use a high discount rate.
Clearly, there is a natural conflict of interests between companies with a large defined benefit pension scheme and the trustees:
The company would like the money to be invested to obtain the highest long-term return without taking undue risks or paying excessive management fees. That way they can minimise their contributions and potentially access the surplus at some future date.
The trustees would like to minimise the risk to their pensions and to themselves. If the fund underperforms due to overcaution on their part then they the company will just top it up. If it underperforms over some period due because they are invested in higher risk, higher return assets then they will be criticised (at best). If it overperforms in the long term then the company takes the surplus but the trustee gains nothing. If there is a choice between a boring index tracker charging 0.1% or a similarly performing high profile big-name fund charging 1.5%, then it is probably safer for them to choose the later. If they are in surplus sufficiently to get an offer to buy out the fund from an insurance company, why wouldn’t they take it, regardless of value for money?
So why would the Norcros trustees choose this (shall we say) high-return asset mix and furthermore plan to keep it? Perhaps they are not confident of Norcros’s ability to fund the deficit and feel it is safer to make it up on their own? Maybe they have a particularly good relationship with Norcros? Either way the situation doesn’t seem sustainable: either Norcros becomes more stable, the trustees change / become more worldly-wise, or we’re hit by a downturn in the equity market.