Prof. Admati, Your Plan Won’t Work
I’m sure there’s a lot to admire about Anat Admati, the Stanford economics professor lately pushing regulators to force banks to add massive amounts of new equity to their balance sheets. But you especially have to stand in awe of her chutzpah. Here’s a woman with no particular expertise in the banking industry (the Times says that, before the credit crunch, she specialized in “complicated financial models”), who’s decided she suddenly knows enough now to recommend radical changes a 500-year-old industry that worked well enough the old way that it managed to play a central role in financing the rise of Western prosperity. All these centuries, and it turns out bankers were doing it wrong. Who knew?
In particular, Admati says banks are too highly levered. In order to better to absorb credit losses, banks’ balance sheets should consist of something like 30% equity, she says, compared to just 5% now. (I’m not sure how she comes up with that latter number; the capital ratios regulators look at are considerably higher.) If banks had been thus capitalized prior to the housing blowup, Admati argues, the resulting credit crunch wouldn’t have happened at all, or would have been much less severe. She’d have banks get to her equity target by restricting dividend payments to shareholders entirely until it was met.
It all sounds very sensible—except that Admati’s scheme would be a disaster, for a variety of reasons. The first has to do with arithmetic. As it is, in the current post-Dodd-Frank environment, the typical ROE at a big bank is around 11%, or not much higher than the typical investor’s cost of capital. That’s with equity at 5% of assets. Now do the math. Boost that 5% to 30% (and tell shareholders they won’t be getting any dividends for the next several years, by the way) and banks’ ROEs would fall to the low single digits. Capital would flee the industry—the very opposite outcome Admati presumably wants. Alternatively, banks might try to boost returns by writing riskier loans. But that would invite the heightened credit losses that have gotten Admati so exercised in the first place. Or banks could raise the cost borrowing. But that would drive more borrowers to the (unregulated) shadow banking system. I somehow doubt that‘s the outcome Admati has in mind.
As I say, a disaster. Nor would Admati’s 30% solution, had it been in place when the housing market blew up, have necessarily prevented the credit crunch that followed. Do you remember what people were thinking during those dark days? The subprime mortgage meltdown was only going to be the beginning, they warned. The consumer borrower was in extremis: after mortgages, the credit quality of auto loans would soon collapse and, after that, credit card loans. It would add up to a massive credit collapse. Commercial real estate would be next and then, who knows? At such a level of uncertainty, no one could know for sure exactly what a given bank’s assets were worth. In the very worst case (which at the time was the only case that providers of overnight funding seemed willing to consider) even a 30% haircut might not be enough. Fear can be funny that way. So that hypothetical 30% equity cushion very likely wouldn’t have prevented the panic from happening.
Admati may not like it, but by its nature, the banking business involves lots of leverage. Deposits are liabilities, remember, and deposit-gathering is a core banking activity. You can argue, and I’ll agree with you, that by the mid-2000s bankers (led by the Robert Rubins of the world) became so lulled by their newly invented imaginary skills at “better managing balance sheet and credit risk” that they operated at levels of leverage that, in retrospect, were plainly too high. But since then, the business has added massive amounts of new capital—to the point that it’s having trouble putting it all to profitable use and is searching for ways to efficiently return capital to shareholders.
Admati seems to want to prevent the possibility of another banking crisis, but that’s the wrong goal. Profitable lending has to entail risk–and as long as there’s risk (and human emotion) there’s going to be the possibility of an institution failing. It’s called free enterprise. What Admati ought to be worrying about, then, is figuring out a way to let a single large institution fail without engulfing the entire system in a panic, and without involving taxpayers. That would solve the problem Admati wants to fix. Boosting equity levels through the roof though wouldn’t fix anything–and would instead drive the banking industry straight into the ground.
What do you think? Let me know!
20 Responses to “Prof. Admati, Your Plan Won’t Work”
Tom, I agree it’s a dumb idea. But the question is where is the risk? If the risk is permitted in the credit assets on which the bank earns a return, then lower leverage is appropriate. If however the risk is largely taken out of bank credit assets, then higher leverage is fine. It seems to me that the problems of the industry in recent years was too much undetected risk in the credit assets. It didn’t used to be that way. Any of us with gray hair remember when bank credit assets were largely pristine. We lost sight of that pursuing our 20% ROEs–remember that?
The risk should not exist both in our assets and within a highly leveraged balance sheet. Choose one place or the other but not both.
So many “smart” people wrongly place blame for the crisis squarely on the banks and ignore the fact that consumers engaged in dubious actions to get a house with an assist from uncle sugar. If I do not understand the cause of a problem I generally offer a solution that will not work and cannot fix the problem. I wonder if the prof has a complete understanding of the cause of the crisis. Only then should she offer solutions which will then be based on sound reasoning an not fairy tails.
Hurray for the Stanford professor. She really has got it right. With enemies like hers, how can she be wrong.
I think you should, perhaps audit some of Anat Admati’s classes. You clearly have a lot to learn.
Mr Zimmerman, please teach me how raising bank capital ratios wont result in high loan rates?
tom
I satrted my banking carreer in 1974 as a commercial lender. During the 1970s and the early 1980s banking was a conservative business with limited risk taking. Commercial customers had strong balance sheets and loans were well secured. There was limited if any leveraged lending and when done was structured on an ABL basis. Not sure about capital ratios but I worked for a top 100 bank that had an 11% tangible capital ratio in 1983. It all stated to change in the late 1980s LBOs realestate oil and gas. Who remembers the Continental failure or that all of the large texas banks failed in the early 90s. The cause -excessive high risk lending. For the last thirty years banking has been a series of blowups caused by uncontrolled risk taking with the last crisis almost taking down the economy. Today capital ratios have not changed much and maybe lower than what we saw in the 1970’s and 80s while banks are carrying more risk and complexity. Today it is much more difficult to find profitable business thus the vulnerability to engage in higher risk activity. Higher capital ratios will mitigate this as there is no way to effectively allocate capital in a risk based manner. The disclosure is not there, Finally we need to stop blaming the consumer and the government for the last crisis. Any one who worked in banking durring that period knows it was the drive for earnings that drove the risk taking
Have been out in the business world since the mid 1960’s and have always relied on one simple premise to be successful.
KNOW YOUR CUSTOMER!
Since then, all the equal opportunity and fairness regulations and policies driven into our business activities by Federal and State do-gooders’ (while some regulations have merit), we lost that which is paramount, KNOW YOUR CUSTOMER!
We business people, especially banks, were forced to go with “check a box” decision making instead of a just basic understanding of what it takes of a customer to meet the obligation being undertaken.
Since we couldn’t use our brains to make decisions it is no wonder that there were “unexpected failures” that just happened to “surprise” us.
Yes, “figuring out a way to let a single large institution fail without engulfing the entire system in a panic, and without involving taxpayers” should be the goal. It does not seems, however, to this former banking reporter that people with “particular expertise in the banking industry” even have much interest in that conversation. One rarely even sees much acknowledgement that something terribly wrong happened in recent years. Many of us would have to stand in awe of the chutzpah in the mostly “Aw shucks, stuff happens” response from the experts.
The argument against increased loan pricing is competition from non bank sources. Car and home loans and credit cards are mostly securitized or provided by non bank lenders. Leveraged loans are sold to loan funds Speciality and niche lenders are entering the market at a rapid pace. None of these sources are subject to bank capital requirements. Whats left? small business middle market loans and real estate and my guess is the market would find a way to fund these loans if bank rates increase beyond a certain point.
Tom, If you want government interference out of your life, then let banks stand on their own, without the backing of the federal government. The reason banks maintain low capital levels is because the taxpayers support their risk taking through FDIC insurance and bailouts when they are on the brink of failure. What the professor is recommending is that banks pay for their own risk taking, instead of the government interfering in their lending activities. Small businesses accept their own risk taking without government support. Why not your treasured bankers?
How long must the industry, commercial banking, wait for some one to do the research and the report on WHO WAS THE INSTIGATORS OF THE PREDICATABLE HOUSING MARKET BLOW UP. TRY THE NOW RETIRED CONGRESS MEN BARNEY FRANK AND CHRIS DODD. Who forced the banks into making uncollectable residential real estate loans to unqualified Borrowers. The loans were going to be charged off as soon as they were made. Why did the Coomercial Bankers allow themselves to pushed into that type of lending with out at leased a fight. One might justifiably say they were forced into it (CRA)and they went along for the ride.
Tom – So I’m a bank with 7% equity and require an 8% rate on that loan. But I change my capital structure so that it’s now 30% equity and I suddenly need a much higher rate? Please teach me why this loan now requires a higher rate?
Tommy H
Ernst & Young’s Financial Services practice in August 2013 wrote a piece on increased capital requirements. The conclusion in response to increased capital requirements in response to Basel III says, “The new regulatory capital rules under Basel III will increase the capital that banks must hold against their commercial mortgage exposures. The result will either be more expensive mortgage rates or less capital allocated to commercial real estate.” The paper goes on to say, “In aggregate, the changes may have the unintended consequence of moving more commercial mortgage lending out of the banking system to the commercial mortgage backed securities market or to a nonregulated environment with non-bank lenders.”
In short, higher interest rates charged due to increased capital requirements, less regulated capital available to borrowers pushing borrowers to nonregulated lenders.
I suppose if E&Y said it then it must be true!
In the 1980’s I was in a bank that Professor Admati would certainly approve of. I believe the ratio of equity capital to assets was about 32% at that time. The older gentleman I spoke to, who was President and Owner of the bank, was clearly proud of the ratio his bank had attained over the years. I imagined the man clearly remembered the Depression and related bank closings and never wanted that fate to befall him or his bank. As I walked out of the bank I saw the town around the bank had died. There was very little business activity taking place and little vision evident for activity to pick up in the future. I am not saying this is cause and effect, but 30 years later I am still struck by the contrast of a bank loaded to the gills with capital and a town apparently starved of it.
DSB – there is absolutely no relationship between the two.
“I am not saying this is cause and effect …” can be found in my comment above. However, all else being equal it pretty much is. At 8% required capital a bank can make a lot more loans than it can with a 32% required capital level – 32% because the math is easy and that is about what the bank I mention had. So if a bank makes multiples fewer loans there are less deserving projects that get funded, built and operational. The projects that never were have associated jobs that never were. The bank making fewer loans needs fewer bankers, analysts, loan clerks and maybe even compliance personnel. These jobs are gone in communities throughout the country.
The bank making fewer loans will charge a higher rate of interest on the loans they make if we are to believe Ernst & Young from the paper mentioned above. Spread income will decline because they won’t be able to charge enough interest to make up for the assets they will shed in order to attain the regulated 30% capital level. Maturing loans won’t be renewed forcing borrowers to seek other bank lenders also shrinking balance sheets, be forced into nonregulated lending markets or forced out of business altogether.
Banks will shed assets because there will be a shortage of investors willing to earn low to medium single digit returns on equity. How else will banks, except for earnings retained over years if not decades, capitalize themselves at the much higher levels of regulatory capital?
Finance is too big a part of the US economy. It needs to be reduced, but Professor Admati’s approach is the wrong way to go about it.
I started to respond to this but it is so riddled with logic flaws and unsupported assertions that a proper response would take too long. Read Admati’s book.
Tommy H – Let’s see if using some real numbers will help you understand the consequences of a sharp increase in the required level of equity in the banking system. The math is easy and the industry information can be found at the FDIC website here https://www2.fdic.gov/SDI/SOB.
Using the 12/31/2013 balance sheet you can see that all FDIC insured institutions had total assets of $14.7 trillion. If you take 30% times $14.7 trillion the Admati required equity is $4.4 trillion. Also on the balance sheet you will find that actual bank equity is shown at $1.6 trillion (higher than 5% of total assets as Tom Brown says). The gap in equity between existing and Admati is $2.8 trillion. To put that number in perspective it is 6.4 times the amount of TARP that was ultimately distributed to banks. The least disruptive way to close the equity gap would be to let the banks earn their way into it. Now we need to go to the income statement to find net income, which was $154 billion for 2013. Some more simple math – take $2.8 trillion divided by $154 billion and you find that it takes 17.98 years of bank net income to close the gap. The 18 years assumes; (1) there is not another recession and (2) the banks will not need to dividend any earnings even though they had dividends of $87 billion in 2013. Neither are realities so it will take longer than 18 years unless you increase interest rates on loans and/or slash costs. Since neither is palatable to you so let’s move onto the next scenario.
The next alternative is for banks to reach out to investors to fund the $2.8 trillion Admati gap. The investors will want to know how much the banks will make. Since you have handcuffed the banking system to not increase interest rates or cut costs the math here is also easy – it is basically $154 billion. The investors will look at the annual net income and divide by Admati equity of $4.4 trillion and see a return of 3.5%. Why would any investor invest in a risky security with a subordinate claim on the balance sheet in order to get what is today the equivalent of a “riskless” return? The answer is they would not so onto the next scenario we go.
If we can’t close the gap in an appropriate timeframe through earnings or raise equity capital from investors, our option is to shrink the balance sheet down to what the existing equity of $1.6 trillion can support. The new smaller balance sheet has total assets of $5.5 trillion and requires that the banking system jettison $9.2 trillion or 63% of current total assets.
I say the consequences of Professor Admati’s plan are significant and would pose a great risk to the economy through the transition to it and beyond. The risks would be much greater than those posed by the Great Financial Crisis, which was scary enough. What does your math show as the way forward?
Let’s assume that your numbers are correct. I have not checked. Admittedly the results are unusual and not something we are accustomed to seeing. Because of that, you immediately dismiss the numbers. But take a step back…
Using your numbers, banks hold $1.6 tri in equity and earn $154 bil for an ROE of 9.6%. The market then prices securities based on that ROE, bank growth AND the riskiness of the banks. Now banks raise $2.8 bil of new equity. Nothing else changes. The result would be ROE’s fall to 3.5%, growth does not change (in Admati’s world, not your jettison assets world) HOWEVER, the RISKINESS of banks also plummets. ROEs fall but risk falls as well.
Some very smart Nobel prize winners discovered long ago that valuations do not change if you change a company’s capital structure. Dust off one of your old finance text books and read for yourself why that’s the case.
If investors have no problem with the returns and capital structure of today’s banks, then they would be equally unfazed by banks operating in an Admati world, without banks selling off assets. The result? An Admati capital raise would be a big non-event.
Comments are closed.