Reading note: Our Lives in Their Portfolios
Brett Christophers | 2025
Introducing asset-manager society
Brett Christophers’ Our Lives in Their Portfolios chronicles the rise of infrastructure as an asset class over the past 30 years.
Christophers comments heavily on the adoption of so-called ‘asset manager society’, whereby large financiers (Blackstone, Macquarie, Brookfield etc) increasingly own what was historically public infrastructure.
The book captures the fluidity between pension funds, institutional investors, asset managers, and SPVs better than I’ve seen in the popular economic literature.
As someone who is paid to work on infrastructure economics issues - the book does a good job at explaining why funds make their equity arrangements deliberately opaque (hint: it’s usually to make their stake easier to offload).
How we got here
A good chunk of the early chapters of the book documents the rise of Macquarie and the privitisation (through concession) of infrastructure in Australia:
Indeed, writing in 2014, Georg Inderst reflected that the Australian financial sector had effectively 'invented infrastructure as an asset class'. This 'invention' occurred in the early 1990s, and was catalysed by a series of privatisations of public infrastructure initiated by both federal and state governments. Electricity, gas and communications all saw significant ownership transformations…
There’s a few economic tailwinds that made this all possible; compulsory superannuation contributions needing to find a home, low interests rates making returns on alternative assets looked attractive, and most importantly, governments looking for ways to finance infrastructure off the balance sheet.
While Macquarie may have kicked things off in the 1990’s, by the time 2009 came around, everyone else had caught on:
But the financial crisis ushered in a radically different macroeconomic era. Western central banks drove down interest rates to close to zero, or even below; and that is where rates remained, by and large, until as recently as 2022. The effect was a dramatic reduction in yields not only on government but also corporate debt. Thus, having come to rely primarily on bonds to earn investment income in the face of the long-term fall in dividend yields, institutional investors were now faced with the novel scenario of not being able to rely on any mainstream financial assets for the generation of recurring annual income at healthy rates of return
In the 20-years to 2009, we also see the rise of renewables as a key regulated asset in just about every economy. All these solar farms, wind turbines, and hydro plants need to be financed, built, and maintained. Who was in the poll position to helo? Asset managers.
Christopher writes:
Owning a coal- or gas-fired plant is not straightforward: the business is a relatively complex and labour-intensive one. Owning a solar or wind farm is very different. Once built and operational, such facilities require very little attention or maintenance, and they have therefore emerged as a favoured investment object for asset managers, who acquire them for their income-producing — not to mention green credential–burnishing — capacity, while outsourcing the little upkeep required to a local contractor.
The rise of housing as an asset class
Christophers mainly points to public infrastructure (think water, roads, hospitals, and maintenance contracts) as the darling class of infrastructure that are usually gobbled up by asset managers.
But since 2008, housing has also been on the agenda. This is especially true in the US:
Furthermore, there was an additional mechanism whereby financial investors could now (indirectly) acquire large stocks of US single-family housing both cheaply and efficiently. The financial crisis saw the aggregation of large numbers of distressed residential mortgages within several different federal-government domains, in particular the Department of Housing and Urban Development (HUD) and the effectively nationalised government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac.
After the crisis, both HUD and the GSEs sold off large numbers of these loans (more than 100,000 in each case) at fire-sale prices, in batches of a scale that only investors such as large asset managers were in a position to acquire. The loans sold by HUD, for instance, were typically priced at discounts of 20–30 per cent on the market value of the underlying property, and 30–50 per cent against the unpaid principal balance. Thus, investors were able to acquire houses by buying these discounted loans, and then foreclosing on the mortgage holder — an opportunity they took up on a large scale.
Accordingly, between 2011 and 2018, the US went from having essentially zero to as many as 300,000 single-family rental homes under the control of large institutional landlords.
Credit where credit is due, the Trump administration has gone to lengths to unwind the attractiveness of housing as an asset class. It’s good policy.
There’s a bunch of ‘social cohesion’ theory here. Home ownership brings with it stability in other aspects of life and community.
A world where your home is maintained to the minimum possible standard (in order to boost valuation of a housing fund) isn’t good for tenants or society at large.
PPPs share risk… right?
Public-private partnerships (PPPs) have grown in popularity since the mid-1990s. The key tenet of a PPP is it allows governments to finance infrastructure and services off the balance sheet.
This quirky accounting treatment is important.
Under most public accounting rules (including in the UK, EU, and Australia) PPP assets don’t hit the balance sheet at all - at least for the length of the concession life. Instead, the government promises a stream of annual payments to the concessionaire.
PPPs only earn off-balance-sheet status if the private partner genuinely carries the risk (thereby not being a liability to the state). As Christophers explains, the extent to which this is true for varying asset classes is hotly disputed:
Christophers summaries:
De-risking of infrastructure investment comes in many guises, and has been initiated by a range of different actors. Two of the main categories of risk that it aims to mitigate are construction and demand risk. The former is the risk that a project costs more or takes longer to complete than projected, or that construction to the required standard is simply unsuccessful; the latter is the risk that, once built, a project does not generate the level of revenue anticipated.
In theory, this is a fair risk sharing arrangement. The sales pitch from an asset manager to a government looks something like this:
First, it said, private ownership maximises infrastructure investment and service quality because public-sector operators ‘are prone to underinvest’ and thereby to create ‘a poor experience for consumers’. Second, private ownership is more reliable, allegedly delivering new infrastructure ‘more often on time and on budget’ than public ownership, and with a lower risk of project cancellation. Third, private ownership is cheaper, both for users of infrastructure (because it entails ‘greater efficiency’ of operation) and for governments choosing between in-sourcing and outsourcing of asset build and operation (because outsourcing ‘results in lower overall life cycle costs’, thus reducing the cost of public procurement). Last, but definitely not least, private ownership, according to the GIIA, reduces risk to the state and, behind the state, taxpayers, insofar as key risks — ‘design and construction costs, delays, volatile market demand, and operation and maintenance costs’ — are ‘transferred to the private sector’.
Herein lies the rub. For state significant infrastructure (energy, water, hospitals, prisons, etc) it’s almost impossible to outsource the entire (especially reputational) risk of the asset not being available or up to scratch.
Thing do go wrong, and often
Christophers does a good job in explaining the incentives and early history of PPP theory. The author goes on to spend the second half of the book documenting the countless examples of governments left holding the bag when things change.
In practice, a bailout (or formally, ‘top-up for unforeseen capital investment’) is also a great moment for an asset manager to sell their stake.
Christopher explains in one example:
The city had also sweetened the deal for the investors by absorbing the risk of unforeseen capital investment. In 2016, the combination of low water usage and 'unexpected repairs' (paid for by the city) saw rates leap by 13.25 per cent. Flush with that increase, KKR — its putative 'long-term' vision now notably absent — began shopping its majority stake late the same year. In January 2018, it achieved the exit it was looking for, selling its position to another asset manager, Argo Infrastructure Partners. Its five-year investment in Bayonne's waterworks concession earned KKR a chunky internal rate of return (IRR) of some 36 per cent.
There’s some absolute clangers of similar arrangements in the book, whereby governments de-risk the cashflows of the asset to their own detriment.
Those costs should not be underestimated. ‘The fiscal costs of demand de-risking can be significant’, Daniela Gabor has observed; asset privatisations such as PPPs ‘can easily turn into budgetary timebombs’. Gabor herself cited numerous examples of such timebombs. One concerned the Sankofa gas development off the coast of Ghana, a PPP between the Ghanaian government and private-sector investors. Under a ‘take or pay’ clause in the contract, the publicly owned Ghana National Petroleum Corporation must purchase a predetermined quantity of gas irrespective of whether it is able to use it. In 2019 alone, the government’s bill for ‘unused gas’ was $250 million, the result of a combination of lack of demand and delays in building the infrastructure needed to offtake Sankofa’s gas. The contract has become a considerable ‘fiscal burden’ for the country.
It turns out 30+ years is a long time
While asset managers love long term investments, they will very rarely hold a PPP equity state for the full concession length. This makes sense once you understand the fund structures asset managers are working with.
A close-ended fund needs to see a return to investors in 5-10 years, so it’s not surprising equity stakes change hands all the time.
Christophers explains in one example:
The KKR investment with which I opened this chapter saw the US asset manager acquire a forty-year concession to run the municipal waterworks of Bayonne, New Jersey, only to sell out a little over five years later. Having been applauded for its ‘long-term vision’ upon making the investment — forty years is indeed a long time — the firm was widely criticised for taking the money and running upon exiting, seemingly prematurely, with a significant profit. Yet the reality is that, once invested, KKR did not have much of a choice. Strictly speaking, it, or at least the fund it had used to make the investment, could not hold the Bayonne concession for the duration.
There’s a more pragmatic element at work here too. Say an asset manager operates with 20% staff turnover annually (as is common in finance). That means the entire organisation is changing (on average) every 5-years.
As people move on, so do priorities and ‘darling’ projects:
Furthermore, given that asset managers typically have relatively high staff turnover, investors can sometimes be reluctant to commit capital for the long term, since the likelihood of the team originally entrusted with their capital still being in place will progressively decline as the months and years go by. Indeed, this is one reason why some institutional investors actually prohibit the allocation of capital to alternative-asset funds without a fixed duration. Last but not least, even the most long-term-oriented investors may on occasion need to access their capital, and it is not always straightforward to exit from open-ended real-estate or infrastructure funds, even after initial lock-up periods have expired. Liquidity is usually generated not by selling assets, but by attracting capital from new investors to replace that of those who are exiting, and this process (and the price at which such new capital can be raised) is notoriously sensitive to market conditions.
Whether or not an asset manager ‘needs’ to sell an equity stake to generate cash, there’s undoubtedly ‘good’ and ‘bad’ times to offload an equity stake. Put simply, when the discounted cash flows look positive, sell:
There are also often significant harmful effects associated with the implementation of the second golden rule of real-asset asset management. To ready an acquired asset for profitable disposal, it is just as important for an asset manager to minimise the costs of operating that asset as it is to maximise the revenues arising from its operation. Assets are valued by the market on the basis of cash-flows, and these represent a net rather than gross measure.
Sell before the ‘big service’ is due
It’s no coincidence these ‘good’ times to sell are typically before major capital spending programmes have hit the books (and can be conservatively estimated in valuation models).
Christophers writes:
If asset managers that run closed-end funds are incentivised to squeeze operating expenditures associated with real-asset investments, they have an even greater incentive to avoid capital expenditures — that is, expenditures that are expected to provide utility beyond the very near term. After all, if the overriding objective is to exit within, say, three to five years of investment, there will in general be no reason to spend money whose returns would be realised wholly or primarily beyond that limited time horizon. It would be economically irrational to do so; and the closer to exit the asset manager edges, needless to say, the smaller the incentive becomes to invest for the long term.
This is just about the oldest trick in the book.
Yet regulation, or a carefully monitored asset management plan can help governments to stay in control:
This is not to say that closed-end funds never countenance significant capital expenditure; they sometimes do. But they typically do so only under special circumstances of two main kinds. The first is where capital spending is externally compelled. The most common instance of this is when asset managers acquire regulated real assets and in the process commit to carrying out certain types and levels of capital investment.
Optionality is more important than you think
Maximising the value of future optionality sounds a bit woo-woo and technical - but governments can and often do become hamstrung by long-term concessions or PPPs.
This is especially true for transport assets (like toll roads) where the PPP contract stipulates a demand floor:
Configuring assets such as transportation networks or energy systems in such a way as to ensure a steady flow of income for private investors is always a matter of physical as well as financial rearrangement. These are, after all, real assets, and their ability to generate cash predictably can depend on all manner of physical manipulations, from restricting rights of way over the land used by solar farms or wind parks to controlling access to tolled bridges or roads. The physical manipulations required to render 'bankable' a particular asset in a particular geographical location can potentially be as impactful for the state's ongoing planning capacity as the financial manipulations discussed above often are for its ongoing fiscal capacity. This is a distinct but often no less significant 'cost'.
That’s one of the benefits of balance sheet financing - governments only have voters to convince if policy priorities change, not an (often adversarial) concessionaire with the means to fund the best lawyers in town.
What is to be done?
So. What is to be done?
Christophers outlines an elegant framing of the problems with PPPs - but is largely silent on what good looks like.
One preliminary question is whether asset managers will own an increasingly large share of public infrastructure over time.
The answer Christophers provides is an emphatic yes:
Cesar Estrada, of the US asset manager State Street, has offered a particularly vivid rendition of this sequential, teleological thesis: The rise of infrastructure in some ways mirrors the growth of the real estate asset class some two decades earlier. Whereas most buildings in metropolitan areas around the world were once owned by corporations, governments or universities, many are now in the hands of pension funds, insurance companies, foundations and other institutions. It’s likely that in just a few years, much of the world’s infrastructure — now mainly owned by municipalities, state and federal governments, and corporations — will also find new homes in the portfolios of institutional investors and infrastructure funds.
Bruce Flatt, chief executive of Brookfield Asset Management, is more succinct. ‘We’re in a 50-year transformation of the infrastructure world’, Flatt confidently told the Financial Times in 2018. ‘We’re 10 years in; we have 40 left to go. By the end of that 50 years most infrastructure in the world will be transferred to private hands.
I’m not entirely convinced of this argument.
Governments, and the voting public are becoming increasingly aware of when they’re getting swindled. This is primarily a function of the IRRs for equity investors being public (or at least-back solvable).
Governments know the price paid, maintenance fees spent, and exit prices for equity stakes - so the returns to asset managers aren’t a mystery. It’s over a matter of time before governments compare the benefits of off balance sheet financing to the enormous returns being manufactured by asset manager society.
A rough scope for a sequel
While Christophers has added a valuable perspective to the PPP discourse, the text feels a little incomplete.
Every author needs to draw a line somewhere. Yet, there’s certainly scope for a sequel (or even a rebuttal) book.
Firstly, it would be interesting to chronicle from the government side just how enthusiastic or reluctant policy makers are to use PPPs. It’s clear PPPs have (massive) potential downsides… so why aren’t these risks talked about more in Whitehall and Canberra?
Secondly, in many cases, PPP models have genuinely reduced delivery risk. Documenting these ‘wins’ along side some of Christophers’ horror stories feels only fair.
Thirdly, some sort of framework for what a low-risk vs high-risk PPP looks like seems easy enough to develop. Not all state infrastructure carries the same risk, so a (even high-level) rank-sorting should be possible.
Finally, including first-hand accounts from treasurers or finance ministers that have tossed-up raising capital through the bond market or running a concession tender would strengthen the discussion. Doing a deal with an asset manager isn’t normally preferred by any government - so knowing which trade-offs make these agreements worth it gives us a glimpse of the future.