The finance industry’s lobbyists have publicly challenged our estimate that the largest U.S. banks receive an annual taxpayer subsidy worth $83 billion. We’re glad for the discussion. Understanding this issue is central to fixing the global financial system.

In an editorial last month, we wrote that recurrent bailouts of the largest financial institutions have given them a unique advantage: They get a break on their borrowing costs, because creditors expect taxpayers to support them whenever they get into trouble. Making some reasonable assumptions and employing research by two economists, Kenichi Ueda of the International Monetary Fund and Beatrice Weder di Mauro of the University of Mainz, we put a dollar value on this too-big-to-fail subsidy.

Although other approaches can produce different numbers, most experts -- including Federal Reserve Chairman Ben S. Bernanke -- concur that the subsidy is significant and should be eliminated. It gives banks an incentive to become big and threatening enough to warrant bailouts, and encourages an oversupply of credit that leads to problems such as the crises in U.S. subprime lending and European sovereign debt.

Funding Advantage

In a March 11 missive, five trade groups representing the banks questioned the existence of the subsidy. Even if it did exist, they argued, it was tiny and the Dodd-Frank financial-reform legislation adopted in mid-2010 did away with it. The brief cited an IMF staff report that estimated large banks’ funding advantage to be much smaller than what Ueda and Di Mauro’s research suggested. It then offered three pieces of evidence that the subsidy has since disappeared -- a briefing by Standard & Poor’s and two studies by economists Ken Cyree of the University of Mississippi and Bhanu Balasubramnian of the University of Akron.

Let’s take a closer look. The trade groups say we should believe the 2010 IMF staff report because it’s newer than the “stale and unreliable” data in Ueda and Di Mauro’s working paper, and because the latter included a disclaimer that it “should not be reported as representing the views of the IMF.” Not quite. Ueda and Di Mauro’s early research, later developed into the 2012 working paper, formed part of the basis for the staff report’s estimate. If the data in the working paper were stale, then the report would be more so. The disclaimer is standard boilerplate for IMF working papers.

When it comes to the S&P briefing, we can only urge the lobbyists to look at it again. They claim it shows that too-big-to-fail banks are paying a premium to borrow. Actually, the briefing is about the difference in yield between the bonds of banks and similarly rated industrial corporations. In no way does this address whether taxpayer support makes big banks’ borrowing costs lower than they otherwise would be.

Finally, the two papers by Cyree and Balasubramnian represent good-faith efforts that the authors readily admit are far from ideal. They found that big banks’ borrowing costs increased relative to those of small banks in the second half of 2010, and attributed the change to Dodd-Frank’s elimination of the subsidy. They employed data on 30 banks, only 11 of which were not too big to fail, a small sample that might have skewed the estimated funding advantages of the bigger institutions. Also, their statistical controls could have missed important factors -- such as the brewing European debt crisis -- that might have had a differential effect on the borrowing costs of the biggest banks.

As Cyree put it: “I can’t tell you that this is strictly due to Dodd-Frank.”

Separate Test

Of course, no statistical study is perfect, particularly when dealing with something as difficult to estimate as the bank subsidy. That said, a paper by three economists -- Viral Acharya of New York University, Deniz Anginer of Virginia Tech and A. Joseph Warburton of Syracuse University -- looked at a larger sample of financial institutions and found that the too-big-to-fail subsidy amounted to almost $100 billion in 2010, the year Dodd-Frank was signed into law. The paper also included a separate test, which looked at bond yields immediately before and after the House and Senate reconciled their versions of the bill. It suggested that Dodd-Frank might have actually increased the subsidy.

Who’s right? Well, given that Dodd-Frank has been significantly delayed and watered down, it’s hard to imagine that the too-big-to-fail problem has been solved. The law’s designation of certain financial institutions as systemically important might even increase expectations of support. U.S. regulators have yet to offer a plausible road map of how such institutions could be liquidated in an orderly manner or set capital requirements that would make them less likely to fail.

What’s at stake is clear. As we said in our editorial, the taxpayer subsidy is lucrative enough to be a major driver of the largest banks’ profits, helping to justify hundreds of millions of dollars in political spending every election cycle. In the absence of subsidies, more shareholders might wonder why banks need to be so large that their executives can’t manage them effectively -- a reality made clear by last year’s multibillion-dollar trading loss in a unit that was supposed to be investing cash and mitigating risk at JPMorgan Chase & Co.

If big banks don’t get a subsidy on their debt, it’s hard to understand why they’re so adamantly opposed to measures, such as increased capital requirements, that would put a limit on their borrowing. Large banks commonly borrow $25 or more for each $1 in equity -- or capital -- they get from their shareholders, compared with less than 50 cents per $1 of equity for the average U.S. corporation. This extreme leverage isn’t the historical norm, and makes the entire system much more vulnerable to crises. As a recent book by financial economists Anat Admati and Martin Hellwig explains, a higher equity level has a negative effect on a bank’s funding costs only if some kind of subsidy has made borrowing artificially cheap.

Requiring banks to have a lot more equity -- we have recommended $1 for each $5 in assets -- would reduce whatever taxpayer subsidy they receive. It would also make the entire financial system much safer. What are we waiting for?

To contact the Bloomberg View editorial board: view@bloomberg.net.