As a consequence, there is a belief that the larger the pension scheme, the lower the cost base.

This assertion may hold because, as I have shown in previous articles, mandate size tends to increase – and the maximum number of mandates tails off – as scheme size increases. In other words, as schemes grow in size, they tend to relatively fewer and increasingly large mandates.

The overall question of whether scale economies exist at a scheme level (and the concomitant pressure to pool schemes to achieve scale) is complex with many different inputs, of which size is only one.

But we can also assess the existence or otherwise of scale economies at the level of individual fund products. Unsurprisingly, the story here is also pretty complex, but a few conclusions emerge.

Bigger not always better

First, economies of scale do exist, but not for all asset classes.

In broad terms, equities tended to show clear scale economies, but trends were less apparent for fixed income, alternatives and illiquid products.

For example, the mean total cost of the largest quartile (by assets under management) of infrastructure fund mandates was 1.6 times that of the mean total cost of the smallest quartile; in other words, no economies of scale but possible diseconomies of scale.

In contrast, the mean total cost of the largest quartile of global active equity mandates was 0.7 times that of the smallest quartile – a clear scale economy that suggests you should buy big to get cheap.

Of course, detractors will immediately complain that this does not consider performance, and assessing funds purely on the basis of cost is wrong.

I will discuss the relationship between cost and performance in the next article in more detail, but the graph shows that there is an inverse relationship between cost and performance in almost every asset class.

Second, for active equity (and possibly other asset classes) segregated mandates show greater scale economies than pooled. The mean total cost of the largest quartile of pooled active equity mandates was only 0.9 times that of the smallest quartile, compared with 0.6 times for segregated active equity mandates.The real issue for small schemes

For some asset classes, the proportion of total costs that are implicit increases as total costs increase (see graph).This means the more you pay in total, the greater the proportion of that total cost is ‘uncontrollable’ or ‘unexpected’ – trading cost in other words.

To put it in slightly more controversial terms, if you sign up to an expensive fee schedule, then there is a probability that you will end up paying an unexpectedly large total cost ex post. This is not going to go down well in some sectors, but the data is clear.

What am I getting at here? From previous articles, you may recall that small defined benefit funds tend to have high allocations to passive equity and use small pooled mandate structures.

In contrast, large funds tend to have high allocations to fixed income and alternatives held in segregated mandate structures. Active equity is more constant across both large and small schemes.

Given this, the advantage of large schemes with certain asset classes is clear: they can use cheaper segregated mandate structures and gain the economies of scale that large mandates bring. So much so, that the data show that schemes with more than £1bn in AUM can access active equity funds at between 70 and 80 per cent of sub-£100m schemes.

Dr Chris Sier is a Founder of AgeWage, chair of ClearGlass Analytics and led the FCA’s Institutional Disclosure Working Group.

This article first appeared in Pensions Expert here; and is re-published with its kind permission