The principle of risk and how we manage it within investment portfolios is one of those boring ideas that has had an outsized impact on the world.
Speaking about boring ideas – have you ever thought about the standardised shipping container? We’ve all seen them – driving past on the back of trucks or stacked in a shipping yard. They’re low-key background noise.
But while international trade certainly happened before the standardised shipping container was invented, it was only after it arrived that globalisation truly became possible. Every port, ship, truck and crane were suddenly able to synchronize around one standard shape, rather than wade through a sea of bags, sacks, barrels and crates.
And risk diversification in modern portfolio management is much the same. International flows of capital happened before, but the principle of risk diversification in portfolio management drove financialisation.
The idea here is relatively simple: you can lower your risk of loss if you invest in multiple investments rather than just one. It’s plain common sense. Don’t put all your eggs in one basket.
It’s such an important concept, and it’s why Markowitz won a Nobel prize for it. Although, as you may guess, I’m about to argue that we need to apply it a little less enthusiastically.
But before I get to that…
How risk diversification works
The theory begins by with two kinds of risk:
- Systematic risk – the general risk of the market having a collective downturn; and
- Unsystematic risk – the particular risk of an individual company going through a negative business event.
At this point, the portfolio management question becomes: how many individual investments do you need in your portfolio before the unsystematic risk is fully diluted and you’re only left with systematic risk? And the Nobel prizewinning answer was “about 20”.
This idea then took off in several ways. Portfolio managers realised that systemic market risk could be further broken down into asset class risk and began diversifying across asset classes. As international markets expanded and deregulation gathered momentum, investment managers could look at diversifying across political risk as well. The neoliberal agenda of the IMF in developing countries had a hardline focus on opening up emerging markets, creating options to diversify even further into high growth economies.
The obvious rightness of risk diversification is now so ingrained into our financial thinking that we no longer think to question it. I used to include myself in that camp.
It’s just a demonstrably proven fact, based on the short span of history that has elapsed since it entered mainstream use.
But increasingly, I have started to question it. Today, I suspect that risk diversification is an economic principle that history will eventually judge as dangerous when applied at scale – much like communism, or Collateralised Debt Obligations (more on these in a moment).
The problem is that we’ve embraced risk diversification as a universal rule, while it’s equally possible that it was only a specific investment approach that worked well at first, but we’ve overdone it.
Are we really mitigating risk?
It’s been a journey of realisation for how I arrived here – mostly, I kept running into realities where risk diversification is exactly the wrong and opposite approach to risk mitigation.
Because risk mitigation is the real goal, right? We just want to lower our risk relative to the return we get. And the leap made by modern portfolio management theory is to say that risk mitigation is achieved through risk diversification. That is a true statement in the developed financial markets where this idea was tested.
Our financial language has extrapolated that statement beyond that original stock market data to almost scenarios where risk exists:
- Risk mitigation now means risk diversification
- Risk now means a concentration of risk that has not yet been appropriately diversified.
This is where things get dangerous, because:
- Risk mitigation can sometimes involve concentrating your risk (eg. “Focus on one thing at a time”); and
- Risk diversification can be risky (eg. “Too many cooks spoil the broth”).
I can find examples of the downside of risk diversification at every level of context:
- At the personal level, it is generally best to build a partnership with one spouse at a time. Diversification of partner risk through multiple partners is an amplification of risk.
- On the family front, families should travel together. It would not make sense to diversify the risk of a plane crash by sending each family member with a different airline. That would amplify the risk of someone getting lost.
- At a company level, chasing too many sales channels means that your offering is fragmented and everyone is underserviced. It is usually better to focus on one or two key sales channels and service them well.
- At a market level, overboard risk diversification leads to massive risk amplification. The Subprime Mortgage Crisis is the case in point – collateralised debt obligations were invented on the back of risk diversification. We’re still paying the price for that misstep.
- At the global level, risk diversification is at the heart of the passive investing movement. When these massive blind portfolios tracking systemic risk by following the market, you get the blind leading the blind in a global-sized echo chamber. This investment process buttresses the status quo of the largest companies. I would argue that you can draw a reasonable link between the risk diversification of passive investing and the slow progress in the fight to slow climate change. When money reinforces the status quo, there is much less pressure on those giant companies to change course.
“Risk amplification”
I don’t know of anyone else using the phrase “risk amplification” in the financial discourse, but I think it makes sense to put that phrase into our risk language.
In my mind, we should have a matrix with risk mitigation – risk amplification on one axis, and risk concentration – risk diversification on the other.
Neutralising the term “risk concentration” will help us see its risk mitigating potential.
Neutralising the term “risk diversification” will remind us that diversifying risk does not necessarily mean mitigating it.
When is risk concentration a good idea?
I come from an immigrant family. At some point in the early years of their marriage, my Greek grandparents left Cyprus on a boat. I doubt that they had very much money. Certainly, they had no credit cards or cellphones, or even a realistic expectation that they would ever return to their homeland. They arrived in Africa, speaking no English, and with only a vague idea of some people they knew living somewhere further inland. They managed their way onto a train in Cape Town or Beira (I am not sure which), and they eventually found themselves in a small farming town in Rhodesia, about 70 kilometres away from Salisbury, the capital city as it was then. They set up a farm shop, and began building a livelihood.
What they did is a form of risk-taking that the developed world does not experience directly anymore. But the refugees and migrants that terrify Conservative and Nationalist voters? They do it in spades.
And what immigrants do to mitigate that risk is they build diaspora communities. In these social structures, the members are bound together by a common background, in the face of the overwhelming common struggles that they experience.
Family-focused trading cultures do this diaspora-building work particularly well: the Greeks, the Lebanese, the Indians, the Italians, the Portuguese, the Jews, the Pakistanis, etc. When I travel in Africa, you see their communal worship spaces (churches, temples, shuls and mosques) all over, typically with a community centre attached.
This is risk concentration in practice, where people centre themselves around one particular community. It is also a mitigation of risk, because the community members aggregate as a nexus of skillsets and professional networks. Today, we ask Google for recommendations – before the Internet, you went to the community priest and he’d know who to refer you to. Because he knew everyone.
Risk Concentration shifts toward Diversification
Over time, as various community members become more successful, they began to build relationships outside of the diaspora community. This grows the community’s general access and influence, as it expands that network.
What strikes me here is that risk concentration and risk diversification appear to have a cyclical relationship.
When a venture is started, risk concentration is risk mitigating. As the venture matures, risk diversification is increasingly necessary.
And in my industry development work, I see this as being an essential insight.
When should we be encouraging more risk concentration?
As you’ll see from my post on Industry Development, I’ve been part of a team building a fish farm in rural Mozambique. Chicoa is an anchor farm for the aquaculture industry there, because it has the commercial scale both to create and import farming inputs. The positive externality of operating in a particular location means that smallscale producers can access their own inputs from us at a marginal cost, rather than requiring them to buy full truckloads at a time.
If an aquaculture industry already existed, there would not be a need for an industry catalyst anchor farm. The industry would already be catalysed.
Investors love this part of the equation. But then they move to the negative part, which is that every other part of the business is a DIY situation. There is no cold chain trucking to rely on. There are no third-party processors to use. Fish distributors and traders don’t yet exist. Artisans for commercial farm construction are not available.
The problem with pioneering is that, by definition, you pioneer in a vacuum. There are very few synergistic connections, if at all. There is no diaspora community of business to belong to. That diaspora community will emerge around you in time, if you survive, but only when you no longer need them because you’ve already done the work yourself.
And it’s here, at this starting point, where mitigating risk requires that we concentrate it. Not that we tolerate it – but that it is actively embraced.
When you look at Impact Fund design, you will rarely find the Investment Committee that says: “This business needs an ecosystem. Go and find a transport company, and a construction firm, and a distribution business, and invest in them to grow alongside our core investee.”
That idea would be rejected as a concentration of geographical risk, political risk, currency risk, and all the other categories. The implicit assumption is that this would be “a bad thing”.
But to me, it is clear that avoiding this risk concentration amplifies the operational risk of the business, which is the greatest concern in these early stages. More than that, it is internally inconsistent – it implies that the fund has a choice to “not” take on the risk. If the investment decision is made, then these synergies are required for the investment regardless. The choice is simply between the business doing all that work alone, or doing it in concert with others. The fund can choose the type of ecosystem that it wants to build, but it cannot avoid the reality that the ecosystem must be build for the investment to succeed.
If we had a more cyclical view of risk in our financial toolkit, it would be obvious that early stage investment funds should look to concentrate risk as a risk mitigation strategy. Because otherwise, you force your investee to be all things at once. That’s the risk that ought to be diversified.
Beyond this, I’d argue that building resilience into the synergistic industries for your core investment has multiplier effects. It is job-creating, which is, in turn, market-creating. And you create revenue centres out of cost centres.
Somewhere in our investment world, we have to allow for dedicated hubs of investment rather than only ever investing in one investment spoke at a time.
And what stops us is this obsession with risk diversification as being better than risk concentration. We break that rule in our daily lives constantly. We ignore it as impractical in our business activities all the time.
We should just make sure that we consider it in our investment mandates as well.