How the market blocks accountability
Friday interview: Dan Davies
I am interested in any study of financial systems that does not entirely depend on the analytical toolkit of economics. Dan Davies brings the concepts of organisational theory to bear on market systems that are often treated as if they were purely transactional, and understandable only in terms of supply and demand. Davies worked as an economist at the Bank of England and as a bank analyst before moving to consulting and writing. He is the author of Lying for Money, about financial crime, and The Unaccountability Machine, which we discuss below. Our conversation covers Milton Friedman, private equity, ESG, executive compensation and much more. This interview has been edited for length and clarity.
Unhedged: What is the core thesis of The Unaccountability Machine?
Dan Davies: It is that managers react to a world that is simply too complicated to manage by creating systems to make decisions for them. This could be a natural and sensible thing to do. But in fact, every time you make a system to take your decisions, you realise that it has an attractive property: that you are no longer personally accountable for those decisions. As a result, people tend to build layers of systems in order to protect themselves. And this ends up having deeply pathological consequences.
Unhedged: You use the term “accountability sink”. What does it mean?
Davies: It’s a metaphor for an organisation of any sort — companies, but government departments do this to a huge extent too — where someone creates a set of arrangements allowing them to divert negative feedback, by acting as if they are not responsible for decisions taken by their rule book or their algorithm. I call it a sink because it’s like a septic tank. You get all of the bad stuff, you put it into the sink, and it builds up there for ages, and then sooner or later it starts to overflow. And all of the problems that you’ve been avoiding for the last 10 years suddenly hit you.
Unhedged: There is a widespread idea that one of the main functions of the market is to hold us accountable as individuals, groups and companies. But you argue that the market can be an accountability sink. How?
Davies: Start from the point that finance is a tool of control. Finance has a comparatively small role in channelling savings and investments in order to produce long-term capital goals, and a really big role as being one of the ways in which we exercise control. This is one of the things where David Graeber was absolutely spot on: the debt relationship as a means by which control is exercised. The best way to think about this is to look at the leveraged buyouts movement in the late 1980s, with KKR and Barbarians at the Gate.
If there’s a company, and you don’t like the way it’s being managed, then you can just run a proxy battle and try to take it over. But you have much more bang for the buck if you simultaneously leverage the company up, because debt becomes a signal that swamps all other decision-making priorities. An indebted company has paying the debt and generating cash flow as one of its priorities; a very indebted company has no other priorities. The way that turns it into an accountability sink is that things that are decisions can be presented as necessities.
So if you were to argue for offshoring production, or closing down plants, or making thousands of people redundant, you would have a debate and you would have to justify your decisions. And that’s psychologically difficult. No one wants to get involved in that. If you are able to say that “this is a necessity because the cash flow has to be generated in order to service the debts”, then you’ve short-circuited that argument. And you’ve done it by taking a previously unindebted but not necessarily aggressively managed company, piled it up with debt, and turned it into a significantly less stable company, or one which has to be absolutely focused on cash flow.
Unhedged: You’ve also written about the idea that exclusive focus on maximising shareholder value can be an accountability sink, even in companies that are not massively leveraged.
Davies: The Milton Friedman doctrine encourages people to pretend that they’ve got an LBO corporate raider looking over their shoulder. The way that this developed historically is interesting. Everyone thinks of Milton Friedman as talking about “a fiduciary duty to maximise shareholder value”. But if you go back to the essay in the The New York Times in 1970, he used none of those words. The title of that essay was “The Social Responsibility of a Business is to Increase its Profits”. “Fiduciary duty”: nine times out of 10 when you hear that phrase, it just means someone’s heard it, and it sounded like a great way to say “it really, really is a duty”. In fact, as anyone who’s been following the litigation relating to Elon Musk knows, fiduciary duty is not a concept whereby any court is going to step in and second-guess the business judgments of someone in charge of a company. You have the business judgment rule. “Fiduciary duty” refers to the fiduciary responsibilities of directors, which are a subset of their overall responsibilities.
Unhedged: What other kinds of considerations, beside profit maximisation, should corporate executives be taking into account? There’s a kind of neatness to the idea that the goal is, always and everywhere, the increase in profits over the long term.
Davies: It’s a way of dealing with an incredibly complicated reality, and trying to give a simple yes/no answer. Because if you say “increase profits”, anyone who has been to business school is going to come straight back and say, over what period? According to what accounting standard? With what trade-off between risk and return?
There’s the old Goldman Sachs proverb of being “long-term greedy”. So if you want to maximise profits over the long term, you might do all sorts of things. You might be able to defend all sorts of courses of action. If you talk about maximising shareholder value, then for a quoted company, at least, you can pretty much tell whether you’re doing well or badly by looking at your share price. You’re getting instant feedback about whether you’re doing the right thing. In practice, this allows the company to be managed by its investor relations department. Which in my view is a pretty crazy way to run a company. But you’ve swapped something that is ambiguous for something that’s got a clean yes or no answer.
Unhedged: I think stock compensation for corporate executives is not a very good idea, for something like the reasons that you’ve just described.
Davies: Stock compensation is pretty bad for those reasons. Stock options are in many ways even worse, because it takes all of the problems of stock compensation and then gives it an additional risk sensitivity to volatility.
Unhedged: I’d also agree that companies that make an explicit goal of getting the share price up, or even of maximising earnings per share, are setting themselves up for problems. You can find plenty of examples of companies that fetishised EPS and subsequently became very bad at doing what they actually do. But I am also suspicious of opening the floodgates to the non-business objectives promoted by the ESG or stakeholder capitalism movements, because those objectives are so amorphous.
Davies: I think stakeholder capitalism is a bit of a vague conceptual mess. You know, you’ve got the three initials ESG. Environmental, social and governance — three things which have almost nothing to do with each other. If these things can be traded against each other, it’s sort of like, you might be allowed a bit of racism if you’ve got really good term limits on board directors.
A lot of stakeholder capitalism comes from the old concept of ethical investment. Back in the good old days when I was an equity analyst you had clients that you would talk to, churches and things like that. It was almost a concept of ritual cleanliness. The Church of England didn’t want to be involved in tobacco. The Methodist Church didn’t want headlines in the Financial Times pointing out that the Methodist Ministers’ Pension Fund had investments in pornography.
ESG and sustainability today is based on the idea of a causal link between the investors spending their money and what the company is actually going to do. The theory of what the link is seems to be missing. Most actual ESG funds seem to be managed in a way that suggests they’d be happy to sell all of their thermal coal to the least ethical people in the market, because then they would be able to give themselves a perfect score.
Unhedged: You use the example of the company that has been piled with debt by a private equity firm and therefore its choices are circumscribed. What range of considerations might it be ignoring? What do you think companies should be paying attention to?
Davies: If we go back to economics before Friedman, and in many ways before the mathematical revolution, there were a lot of different theories of the company. The one that really interests me is Herb Simon’s. He said that profits and shareholder returns are not the goal of a company, but a constraint on the company, among other constraints like labour costs, raw material costs and legislation. The company has to provide enough profitability and dividends to keep the investors content. After that, its focus is to just explore the space of possibilities. You develop new products, make long-term investments, and continue going forward as an entity.
Unhedged: Have there been successful companies managed with that more spacious philosophy in mind?
Davies: Historically, the vast majority of companies were managed in this way. Boeing before the McDonnell Douglas merger was a company that existed to make passenger aircraft, among the best in the world, and to make weapon systems for the United States Air Force. You can go back through almost all of the companies that have been around for longer than 50 or 60 years, and you can see them gradually changing their mission statements from, to take the example of DuPont, “better living through chemistry”, to something that keeps emphasising shareholder returns.
Unhedged: I always wonder if profit, like love, is intrinsically a side effect. It’s one of these things where if you aim directly at it, you’re doomed to miss.
Davies: Yes. it’s very similar to the “oblique approach” which John Kay wrote a really good book about. But I was always a financial sector guy, so I come back to things like the Goldman Sachs “long-term greedy” concept. Whenever people start talking about how much Goldman Sachs pays in bonuses compared to how much it pays to shareholders, I think: “I didn’t see any shareholders working 80 hour weeks to generate those results.” One of the most interesting businesses books I’ve read in the last 20 years was from Maurice Greenberg, who led AIG. He said AIG demands a huge amount from its executives. It requires them to put in extraordinary efforts and move their families around the globe. And consequently, AIG would never have redundancies as a means of managing the business cycle, because he thought that was the other side of the bargain, and if he were to start kind of just cutting that would violate the bargain.
Unhedged: You have a very particular way of analysing corporate management, using the metaphor of a kind of very slow artificial intelligence.
Davies: The idea that companies are artificial intelligences keeps on getting reinvented by different people at different times. I found what I think is the most sophisticated overall treatment of this idea in the work of Stafford Beer, who was a management theorist of the 1960s and 70s. He decided that you can apply the mathematics of information theory to organisations in the same way that economists apply the mathematics of constrained optimisation to markets. So the thing you need to do when you’re analysing a decision-making system is to look, at every stage, whether the bandwidth of the management is able to handle the amount and variety of information their environment is generating. Then you start seeing whether a decision-making system is going to remain well regulated, or whether it’s leaking information and is about to spiral out of control.
In the case of companies and markets, the point at which the information starts to leak is almost always in the accounting system, because the accounting system is very much a reduced resolution representation of business reality. So that’s usually where problems start, and management decision-making starts losing contact with reality.
Unhedged: Can you give us a concrete example of how loss of resolution in an accounting system leads to bad decisions?
Davies: The big one is always outsourcing, and the tendency of accounting systems to allocate fixed costs to things like manufacturing operations, and consequently to make it look like outsourcing or offshoring those manufacturing operations is going to save more money than it actually will.
It’s very difficult to get an idea of how much money is saved by outsourcing. One thing you do know is that when you outsource a manufacturing process, you also outsource all of the information about the process. All of the knowhow, any further improvements, any kind of further uncertainty takes place outside the boundaries of your company. And the accounting system will never tell you anything about that, because all of the information about what’s going on in manufacturing is now communicated to you through a sense of numbers going back and forth. And that’s one way in which I think companies, but also government departments who start outsourcing core operations, start losing capabilities and information.
Unhedged: I want to circle back to private equity. The meat of the American economy is being purchased by private equity. We only pay attention to the very large buyouts, but every widget maker in Ohio is now private equity owned. If what you say about management incentives is true, we have a problem on our hands. What do you propose we do about it?
Davies: Private equity changes the incentive structures to the extent that debt is overused, and to the extent that management based purely on financial accounting is overused. And so if we kind of want to attack the problem, we want to attack those, rather than the ownership structure itself. It’s hard to ban private equity because if someone thinks they can run a company better than the incumbent management, they’ve got a right to buy shares.
Something that could be worth looking at, which relates to an idea that Paul Volcker was proposing towards the end of his term in the 1980s, is to look at the extent to which limited liability should be available to investment vehicles. The history of limited liability was that it was a way of getting the savings of the middle class into investments, into corporate goods. But once limited liability is applied three layers deep, between an investment fund, its feeder funds, and the underlying investors, it becomes less obvious that is the same kind of implicit bargain we made in the 1820s, when the joint stock corporation was first proposed.
Unhedged: How might tempering the scope of limited liability change organisational behaviour?
Davies: It very much changes the incentives to make use of debt financing. Because if you’re a private equity fund manager, you might as well lever up all your portfolio companies with as much debt as they can take, because if one of them goes bust it doesn’t affect the rest of the companies. It doesn’t affect the rest of the returns on the fund. If you find yourself having to guarantee the liabilities, including the pension liabilities, of every company you take over, then all of that portfolio company leverage becomes fund leverage.
Unhedged: Any final comments?
Davies: The only other thing I’d say is that, for governments, the market is the ultimate accountability sink. Whenever you want to make a genuinely unpopular decision about taxation, spending or pensions policy, you blame it on the market. And that’s the biggest single use of decision-making systems to avoid making decisions. And I think that that’s at the root of a lot of populist politics we’ve got. It just becomes, in the end, intolerable to the people who are the subject of the decisions. And that’s the trouble with accountability sinks. They store negative feedback, but don’t deal with it. And then suddenly, you get years and years worth of negative feedback suddenly released.
Recent Comments