Lack of hard evidence behind LGPS reform proposals
Anyone who listened to the 4th December LGPS Live webinar where MHCLG explained the proposed reforms will be in no doubt of the Government’s determination to push ahead with consolidation within the LGPS. I couldn’t help but notice that the slides had the Treasury’s logo displayed more prominently than MHCLG. It looked deliberate.
The general view is that the consultation reads well, and that the proposals are for the most part coherent. It feels almost as if one of the pools has reviewed it before publication, as it contains fewer than usual examples of lack of understanding of how the LGPS really works.
That said, there are many areas, some more obvious, others more nuanced, where funds responding to it will want to use the opportunity to sway the Government’s mind. Examples range from the timescale, described as ‘challenging’ by MHCLG in the webinar, to the decision to give pools’ primary responsibility for advising funds and the practicalities of the proposals for local investment.
A common reaction has been to point out the lack of hard evidence behind the reforms the Government wishes to make. That applies both to the drive for scale through consolidation, and also to the handing over of so much of the responsibility to the pools before they have developed the capacity to take it on.
In this article, I review three areas: one is the net value added to investment returns by similar funded public sector pension funds round the world (the Chancellor’s favoured measure); the second is cost savings; and the third is the lack of any validation that alternative ways to achieve the Government’s objectives are not superior. My conclusion is that, despite the Government’s determination to push ahead, there is no evidence to support these reforms.
Misleading evidence needs to be called out
In one specific case it is worse than that. Misleading evidence about cost data appears to have been given to the Department of Work and Pensions Select Committee into DB Pensions. If that data lies behind the determination to push these reforms through, then it is important that it is called out now.
Many LGPS funds find themselves for political reasons unable to say what they really think about these reforms, and will wisely choose to pick their battles when responding. But making decisions as important and disruptive as these reforms without any hard evidence is clearly to the detriment of both the more than six million members of the LGPS and employers. We need to have a good rationale for believing the benefits will be greater than the disruption. That is what is lacking here.
In 1867 J.S. Mill said, “Let not any one pacify his conscience by the delusion that he can do no harm if he takes no part, and forms no opinion. Bad men need nothing more to compass their ends, than that good men should look on and do nothing. He is not a good man who, without a protest, allows wrong to be committed in his name, and with the means which he helps to supply, because he will not trouble himself to use his mind on the subject.” In that spirit, as a knowledgeable commentator who has no qualms about saying what I believe, I am setting down my protest. I am not against consolidation or pooling per se, but only against it being done badly.
Mega-funds have not achieved better returns
The Government has categorised the LGPS as inefficient and fragmented. The Chancellor has looked at the Canadian funds as examples of how mega-funds might operate more efficiently. However, the Canadian funds have not delivered better outcomes. The table below shows the actual and benchmark 10 year returns (i.e., the Net Value Added) of six Canadian schemes alongside those of the LGPS.
Net Value Added over 10 years: 6 Canadian superfunds vs LGPS
Actual | Benchmark | NVA | |
CPPIB | 10.0% | 9.2% | 0.8% |
CDPQ | 7.4% | 6.5% | 0.9% |
Ontario | 7.6% | 7.5% | 0.1% |
PSP | 6.7% | 7.2% | -0.5% |
BCI | 7.8% | 7.1% | 0.7% |
Alberta | 7.3% | 6.9% | 0.4% |
LGPS | 7.3% | 5.5% | 1.8% |
Sources: annual reports, SF3 data; LGPS benchmark = avge discount rate +1%
Of course, the LGPS has no single reference benchmark. I have therefore used the average discount rate over the past ten years (4.5%) and added on 1% to reflect the fact that most committees choose to take more risk when setting their asset allocation in order to keep contributions down. Readers may choose to add a higher number and reduce the estimate of the LGPS’ NVA.
Two conclusions are clear. One is that the NVA of the Canadian funds has been modest, largely between zero and one percent. The second is that the LGPS has delivered at least as much and probably more NVA. As context, I also looked at some other funds in Sweden, Finland, New Zealand and Australia. The story is similar there. The New Zealand Superannuation Fund, the smallest, added most NVA.
Mega funds are not cheaper
The potential for cost savings has been a standard trope when government has discussed LGPS reforms. We were told that savings from pooling so far amounted to £320m up till 2023 when SF3 data (to March 2023) suggests that investment expenses actually rose by £680m over the preceding five years. These are not mutually exclusive statements: it is quite possible that costs would have risen further without pooling, but the crucial point is that no evidence has been provided.
I therefore also compared the same Canadian funds’ costs with the LGPS as reported in their latest Annual Reports. LGPS data comes from the most recent SF3 data to March 2024.
2023 investment costs: 6 Canadian funds vs LGPS
Internal | 3rd Party | TOTAL | |
CPPIB | 26 | 69 | 94 |
CDPQ | 19 | 40 | 59 |
Ontario | 34 | 41 | 75 |
PSP | 26 | 40 | 67 |
BCI | 20 | 61 | 80 |
Alberta | 16 | 51 | 66 |
LGPS | n/a | n/a | 49 |
Source: annual reports, SF3 data to 3/24; LGPS includes governance costs
I have included governance costs in the figure for the LGPS, as some pooling and similar oversight costs will have gone into the category of governance costs. The 49bps figure is still almost certainly under-reported, as not all funds include third party management costs charged directly to the NAV of the investment vehicle as investment costs. But even if we take the 25th percentile fund within the LGPS, its investment costs were 69bps. My point is that the evidence for aggregate cost savings through creating mega-funds is tenuous at best.
Misleading evidence presented to the DWP Select Enquiry into DB Pensions
It gets worse. Evidence into LGPS costs which was actively misleading was presented to the DWP Select Enquiry into DB pensions in May 2023 and used as a core argument to consolidate the LGPS.
The Pensions Insurance Corporation (“PIC”) submitted written evidence (DBP0073), in which they argued that the one case for mandatory consolidation was the LGPS. The evidence they produced to support this was to compare the 2022 costs of the Canada Pension Plan Investment Board (“CPPIB”) with the LGPS. They state (paragraph 44) that:
the CPPIB costs £1billion a year less to run than the LGPS. This £1 billion represents a direct transfer from local taxpayers struggling with a cost of living crisis to the myriad advisers and service providers employed by each of the 100+ LGPS trustee boards
and (paragraph 46):
The LGPS pays 56bps per annum in fees, up from 45bps in 2017/18. This compares to 27bps per annum for the CPPIB12.
These statements to the Select Committee are demonstrably erroneous. The CPPIB 2022 Annual Report (p52) gives total investment costs of CAD 5,967m (approx. £3.3bn) on assets of CAD 539bn. PIC has included in its evidence ‘operating’ expenses i.e., internal CPPIB costs, but failed to include third party management or performance-related fees.
They compare this with LGPS data which does include third party costs. Their LGPS data is also erroneous, though to a lesser extent, as they have included administration costs. If we exclude those, the ‘correct’ LGPS number as reported would be 52bps.
In reality, CPPIB investment costs in 2022 were approximately £1.5bn higher than the LGPS’s at about £3.3bn, not £1bn lower, and the cost in bps was 111bps. I do not know how much the Government has relied on this specific piece of evidence when deciding to push for consolidation reforms in the LGPS, but it needs to be called out before they present the Pension Bill to Parliament.
Why no public procurement?
Public procurement can sometimes get a poor press as being clunky and ineffective. But it serves an important role in ensuring that public bodies get at least some pricing tension. The LGPS funds are public bodies and subject to procurement rules. I find it hard to understand why they are being asked to transfer important functions such as strategic allocation advice and implementation without going through some form of public competition. The pools by virtue of their recent establishment have limited track records, and in most cases do not have the capability or resources to provide the functions which the Government wants them to take over.
Public procurement, whether on a one-off or a more regular basis, would channel the LGPS funds towards the pools who offer better value. I’d like to see private sector fiduciary managers allowed to compete too, because then we would have some objective evidence whether or not these pooling proposals are indeed in the best interest of members. If the procurement were set up as a framework agreement, the Government could stipulate both the desired legal structure and the number of successful tenderers. That would also tick a couple of their boxes: a smaller number of pools and FCA authorised entities.
The emperor has no clothes
I find myself in the position of the boy who points out that the Emperor has no clothes. There is little or no evidence to argue that consolidation will lead to better outcomes for the members who we owe fiduciary duty to, or for the employers who will be first in line to pay any extra costs. And yet funds are being asked in the consultation to agree to statements that it will. I am particularly concerned that the Government is relying on the misleading PIC evidence to the DWP Select Enquiry.
I shall publish my own response over the next few days. Many may feel that they cannot make the points which I am making here. It therefore falls to the little boy to point out the obvious.
Comments