In two recent posts [1,2], Scott Sumner disputes the role of financial rent extraction in increasing inequality. His best argument is that due to competition, government subsidies by themselves cannot cause inequality. A few months ago, Russ Roberts asked a similar question: "If banking is a protected sector that the government coddles and rewards, why doesn’t competition for banking jobs reduce the returns to more normal levels?" This post tries to answer this question. To summarise the conclusion, synthetic rent extraction markets are closer to an 'Ultimatum Game' than they are to competitive "real economy" markets.

Scott brings up the example of farm subsidies and points out that they only reduce food prices without making farmers any richer - the reason of course being competitive food markets. In my post on inequality and rents, I used a similar rationale to explain how reduced borrowing costs for banks in Germany (due to state protection) simply results in reduced borrowing costs for the Mittelstand. So how is this any different from the rents that banks, hedge funds and others can extract from the central bank's commitment to insure them and the economy from tail events? The answer lies in the synthetic and rent-contingent nature of markets for products such as CDOs. The absence of moral hazard rents doesn't simply change the price and quantity of many financial products - it ensures that the market does not exist to start with. In other words, the very raison d'être of many financial products is their role in extracting rents from central bank commitments.

The process of distributing rents amongst financial market participants is closer to an ultimatum game than it is to a perfectly competitive product market. The rewards in this game are the rents on offer which are limited only by the willingness or ability of the central bank to insure against tail risk. To illustrate how this game may be played out, let us take the ubiquitous negatively-skewed product payoff that banks accumulated during the crisis - the super-senior CDO tranche ((In some cases, the super-senior itself was insured with counterparties such as AIG or the monolines making the payoff even more negatively skewed)). In order to originate a synthetic super-senior tranche, a bank needs to find a willing counterparty (probably a hedge fund) to take the other side of the trade. The bank itself needs to negotiate an arrangement between its owners, creditors and employees as to how the rents will be shared. If the various parties cannot come to an agreement, there is no trade and no rents are extracted.  The central bank commitment provides an almost unlimited quantity of insurance/rents at a constant price. Therefore, there is no incentive for any of the above parties to risk failure to come to an agreement by insisting on a larger share of the pie.

In a world with unlimited potential bank stockholders, creditors and employees and unlimited potential hedge funds, the eventual result is unlimited rent extraction and state bankruptcy. The only way to avoid inequality in the presence of such a commitment is for every single person in the economy to extract rents in an equally efficient manner - simply increased competition between hedge funds or banks is not good enough. In reality of course, not all of us are bankers or hedge fund managers.  Nevertheless, it is troubling that the evolution of many financial product markets over the past 30 years can be viewed as a gradual expansion of such rent extraction.

Although I've focused on synthetic financial products, the above analysis is valid even for many of the "real" loans made during the housing boom. In the absence of the ability to extract rents, many of the worst loans would likely not have been made. The presence of rents of course meant that every party went out of their way to ensure that the loans were made. It is also worth noting that although I have explained the process of rent extraction as a calculated and intentional activity, it does not need to be. In fact, as I have argued before [1,2], rent extraction can easily arise with each party genuinely believing themselves to be blameless and well-intentioned. The road to inequality and state bankruptcy is paved with good intentions.



Another great post! Thank you! I do wonder, however, about the degree to which the ability to extract rents in the real loan market was affected by the availability of an infinite supply of synthetic loans. In a world without synthetics would prices have adjusted fast enough to prevent the development of markets for the worst of the loans? (Note that I'm not claiming to know the answer to this question.)


Carolyn - Thank you! The question of how synthetics affect the real market in this case is a good one. Unfortunately I have nothing smart to say about it! Clearly if there is a large side-market that trades at unreal levels given that the main point of the market is rent extraction, then the real market is not immune. But how exactly the two feedback onto each other is a question I need to think a lot more about.

Diego Espinosa

Ashwin, Thinking of subprime lending, I think you did see a decrease in rents to the underwriting function as competition increased -- this is probably the type of thing Sumner is thinking of. Subprime underwriters reacted to margin declines by holding on to securitizations and accessing leverage to raise portfolio returns. In effect, they substituted rent from the Fed put for eroding rent from the origination oligopoly. As portfolio spreads declined (again, due to competition), higher and higher leverage/ illiquidity was needed to extract the same level of profit, in effect raising the value of the Fed put. As the put value rose, so did the contingent liability to the put provider -- something the Fed never really seemed to grasp. Going forward, an important question is, "how reliable is the Fed put?" We see signs everywhere that markets are regaining confidence in the commitment, even after the Fed's failure to deliver (completely) in 2008. Can the Fed deliver the next time? Inflation risk may yet turn out to be the next thing to tie the Fed's hands. So the consequence of rising inflation is not just to directly raise the contingent cost of holding bonds, but also to lower the value of the embedded Fed put.


Diego - Great comment. The point about extracting more rents from the Fed put as oligopolistic rents declined is a good one - probably even relevant on a longer time scale in a broader sense. Increased competition in combination with the Fed put makes things worse, not better. You may be right. My feeling is that the market views the failures in 2008 as having made it even more certain that 2008 will not be allowed to repeat itself no matter what. Of course, significant inflation will push the Fed's hand but it may act like the BoE which is simply to ignore higher inflation unless it gets well above target. I have a half-written post which I may post soon on this topic - the interaction of monetary policy and the Fed put. The long-run result in my opinion is the monetary policy lever as some kind of broken shower where the only options are too hot and too cold and the Fed keeps alternating between the two.

David Pearson

I'm looking forward to seeing that post. The market does seem willing to believe the Fed on the surface. However, the proximate cause of the financial crisis was the Fed's failure to make good on the original Greenspan put. Something tied the Fed's hands in 2008. It was likely a combination of three things: denial (that they ever made the commitment); fear of un-anchoring inflation expectations; and political constraints. The first one of those is gone -- now they realize they are on the hook. The second one is much diminished, as no inflationary expectations resulted (yet) from their actions. The third is still very much in force. My sense is that if the second one manifests itself, the market's confidence in the put will be shaken. A put that can only be exercised with ultra-low inflation expectations is one that is way out of the money.

Michael Strong

When I posted this to my Facebook account, I received the following reply from Arthur Breitman; I'm curious to know how you would respond. He first quotes you and then the remaining comments are all his: ‎"The bank itself needs to negotiate an arrangement between its owners, creditors and employees as to how the rents will be shared. If the various parties cannot come to an agreement, there is no trade and no rents are extracted." If I understand the argument correctly, this is subtly begging the question. "Employees" are presented as one actor in an ultimatum game. However, not only are there many, many, employees, there are also people outside of the bank willing to work in finance for the right price. Assuming this away is like assuming the lack of a competitive labor market. I am more than willing to agree that financial institutions are not competing on a free market, but that has little to do with the underlying labor market. I think a more convincing and simple explanation of high salaries in the presence of subsidies would be to posit that the labor market in finance is very inelastic.


Michael - That's a good question! First of all, let me clarify that the aim of this post was to establish why even a perfect market with infinite number of all participants doesn't help matters. Obviously I'm not denying that there may be specific skills to banking and various other "imperfections" that may be relevant. Let me try and explain with an example as to why labour market "competition" would not help. Clearly there is no shortage of potential bank shareholders and creditors - even you and I hold their stock and buy their bonds. In this world, let us assume that we have 1000 new entrants into the labour market with the quantitative and finance skills to work in a bank. What I'm saying is that 500 of these entrants will become traders in a bank, 500 will join hedge funds and they'll trade amongst each other. The end result is simply an ultimately higher bill for the taxpayer. The basic fact that the rents that can be extracted can be increased by financial "ingenuity" means that nothing in this process is "competitive" in the economic sense. Now obviously in the real world, there are various elements of competition in the short run. But in the long run, the dominant trend is simply that more and more rents get extracted as more and more people become bankers and hedge fund managers.


David - The post is in the works so hopefully soon. On 2008, my view has always been that if they'd let Bear Stearns go, we'd have had a much smaller crisis. The Bear bailout gave the wrong signals and the Fed and Treasury then tried to backtrack by letting Lehman go with catastrophic consequences. Anyway, at the risk of summarising my future post, the banking sector has self-organised in a manner such that they cannot collapse without risking deflation. All the banks have the same positions so there is no diversity in the sector, the end result of which is a kind of binary state - either healthy banks + inflation or bankrupt banks + deflation.

Bruce Wilder

Why does "competition" to enter markets in some fields seem to drive prices and costs up, instead of down? This is not a phenomena restricted to financial services -- it's seen in health care, as well, which to my mind, suggests that information asymmetry -- common to both -- may be a critical factor. Kevin Phillips in his political study, American Theocracy (2006), had a lot of insight into the details of how this rent-seeking played out with mortgage markets, regulation and Fed policy. When he followed up with his Bad Money (2008), he invoked Gresham's Law (Bad Money drives out the Good) as a metaphor for the deleterious effects of financialization.