Achieving Scale in Pensions Through Portfolio Aggregation

Clive Lipshitz sent me a guest comment on achieving scale in pensions through portfolio aggregation:

Last year, Ingo Walter and I published a comparative study of the Canadian and U.S. public pension systems. We describe reforms undertaken in Canada in the 1980s and 1990s and illustrate the positive impact of these decades later. To do this, we held extensive discussions with pension experts including provincial finance officials, actuaries, academics, and pension CEOs. We supplemented this qualitative input with a novel dataset with tens of thousands of observations from public filings of the largest U.S. and Canadian plans to compare outcomes in the two countries.

Drawing on lessons from Canada, we detail a series of proposals to improve U.S. pensions through better design and structure, enhanced governance, closer alignment of assets to liabilities, and ultimately better funding levels. We document the truly distinguishing features of Canadian plans, many of which have nothing to do with investing (they relate mainly to joint sponsorship, governance, and plan design). That said, certain investing features are significant. One of them – while not unique to Canada – is evident across the country. By aggregating investment pools, Canadian plans enjoy substantial economies of scale.

Subsequent to publication of our study, we promulgated our recommendations in a number of publications, including the Journal of Portfolio Management, The Hill, Institutional Investor, and MarketWatch. Moreover, I have engaged a range of U.S. pension stakeholders, including state treasurers, pension trustees, pension CIOs, and asset managers. Aggregating investment pools is an idea that resonates.

Historical context is necessary to understand the Canadian structures. Pre-reform, pension reserves were managed within provincial finance departments and invested almost exclusively in government debentures. Quebec is the exception, as CDPQ had been investing in public markets since 1967 and in private markets since 1971. When the Ontario pension funds were reformed, each established its own investment office – thus were borne the OTPP, OMERS, and HOOPP investment organizations. In the other provinces, investment offices were decoupled from government into distinct Crown corporations and mandated to operate as centralized investment offices serving pension funds and other provincial capital pools as clients. BCI in 1999 and AIMCo in 2008 were established in the western provinces, PSP Investments was established in 1999 to manage post-2000 federal employee pension liabilities. In 2016, IMCO was founded to manage reserves of smaller Ontario plans. CDPQ, preceded all of these. Manitoba has also explored consolidated management of its pension pools.

Scale through aggregation allows smaller pensions to enjoy many of the benefits of the large Ontario plans. These include the ability to establish extensive investment operations – staff and systems – which allow for implementation of sophisticated total portfolio management strategies. Scale enables direct investing in companies, properties, infrastructure assets, equities, bonds, and loans. It also facilitates a different way of engaging fund managers – larger commitments, access to funds that might otherwise be inaccessible, reduced fees, and access to co-investments and separate accounts.

We find that there is a correlation between portfolio size and investment performance. We argue that benefits likely begin to accrue above a certain minimum size. Moreover, Keith Ambachtsheer and his coauthors find that larger pension plans capture more of their total returns by way of lower total expense ratios than those of smaller plans.

What does this mean in the U.S. context? There are more than five thousand distinct public sector pension plans sponsored by states, counties, cities, and special districts. This balkanization is the inevitable consequence of there being so many distinct jurisdictions across the country. As a result, there are thousands of elected officials, civil servants, and worker representatives who sit on pension boards, just as there are thousands of individuals in positions of responsibility for managing pension reserves. Each of these has a position of authority over pension assets. While assets are highly concentrated – the 25 largest plans account for more than half of all pension assets – the vast majority of pension plans are actually very small.

Some states – such as Massachusetts – operate centralized pension investment organizations. Some are actively exploring this. Most are not. We propose that pension systems explore the benefits of scale through consolidated investment organizations or by allocating to portfolio aggregators.

How might this be achieved? Larger states might merge the investment offices of all or some of their various public plans. Smaller states might consider aggregating investment offices across state lines. Newly formed public sector entities might be established for this purpose.

Alternatively, scale benefits might be achieved by allocating to pension specialists such as units of insurance companies or asset managers. Again, there is precedent in Canada in the form of CAAT Pension (which evolved from its origins as a pension provider for Ontario colleges) and Trans-Canada Capital (formerly the pension manager of Air Canada). These are seeking to take on portfolio management for smaller pensions (and total pension management, in the case of CAAT).

If total portfolio consolidation is a bridge too far, then benefits of aggregation could be achieved within distinct asset classes. This is the role IFM plays for a consortium of Australian superannuation funds. U.S. pension funds seeking to obtain the benefits of scale in real estate or infrastructure investing might establish or acquire cooperatively owned asset managers to allocate on their behalf in these asset classes.

Any proposal to do something new requires a catalyst for action. The funded status of U.S. pensions has recovered significantly in 2021. This owes much to tailwinds from very supportive capital markets. It is likely a one-time gain. Ultimately, for this to be lasting, pensions should seize this opportunity to look at plan design, governance, and permanent enhancements to the management of pension reserves, including the benefits of scale.

This is an excellent comment from Clive and I thank him for sending it to me so my readers can read his great insights.

Clive also sent me an email stating CalSTRS’ board’s investment committee recently  reviewed its annual cost report. They’ve been using a “collaborative” approach for many years - it’s about trying to borrow from the Canada model of direct investing, co-investing, and SMAs (see the chart below ).

According to Clive, "it shows that they saved $195mm in management fees and $113.5mm in carried interest last year through this model. It’s not really about aggregation of investment portfolios, but it is about the benefits of scale and of shifting to more of an internal model."

Anyway, I will keep my comments brief, I totally agree with Clive,  US pensions need to amalgamate and attain scale and they need to adopt the Canadian governance model.

Smaller plans like UPP in Canada can still succeed but that's because they hired the right CEO and are adopting the same formula of success as the larger ones, ie. more internal management, more co-investments to reduce fee drag.

In the US, there are many smaller public pensions that need to amalgamate with the larger state ones to  benefit from scale and better governance.

That's all from me, I thank Clive for sharing his insights.

Below, CNBC's Rick Santelli joins 'The Exchange' to report on the sale of 30-year Treasurys.