Interest rate risk, credit risk and a comment on bank margins

A theme of this blog is the interest rate risk and credit risk of American banks. As regular readers know I got the interest rate risk versus credit risk wrong of late.

But being wrong in the past won’t stop me trying to be right in the future. Besides it is worth considering what proportion of a bank’s margin comes from accepting interest rate risk, what portion comes from accepting credit risk and what portion comes from just servicing customers.

The current yield for interest rate risk

Currently you can buy an “on-the-run” Fannie Mae guaranteed 30 year fixed rate mortgage with a yield of 615bps. Before the Fannie Mae credit crisis it was still almost 600bps. Bloomberg gives a series of what the on-the-run mortgages yield – the so-called “perfect coupons”. You can find this sequence by typing MTGEFNCL on your Bloomberg – or just find a graph here.

Currently the intended Fed Funds rate is 200bps and (highly) secured borrowings will cost only (say) 50-100 bps more than Fed Funds.

Now if we presume that the Fannie Mae mortgage is guaranteed by the Federal Government (a good assumption after last week) and is hence riskless then you can earn approximately 300-350bps by taking only interest rate risk. All you do is borrow floating to buy “perfect coupons”. And you don’t hedge anything.

300bps levered 10 times and with 30% tax taken out will still give you a 20% post tax return on equity.

Unfortunately you take a shocking amount of interest rate risk to get that 20 percent return on equity. The mortgages will (at least in normal times) refinance if rates fall. They will however extend for an unknown but long period if rates rise. Pretty well all movements in interest rates are negative for someone taking that sort of interest rate risk. Indeed it is very easy to model insolvency for such a company.

It was my contention – wrong it seems – that the main risk taken by US banks was interest rate risk. You can see the gory details of my mistake here.

State of the banks

There are banks that take mostly interest rate risk and banks that take mostly credit risk.

Banks that take mostly interest rate risk tend to have floating funding and have assets that are either GSE securities or very secure mortgages. Extreme examples included Commerce Bancorp (something I was short on interest rate risk and lost), and New York Community Bancorp. [State Street also takes a surprising amount of interest rate risk whereas Bank of New York tends to prefer credit risk.] Banks that take mostly interest rate risk have had great relative performance. In the event of an inflationary spiral those same banks would under perform massively. [The jury is still out on inflation – another theme of this blog.]

Other banks however eschew almost all interest rate risk keeping their asset and liabilities of matched duration – which usually means keeping securities of short duration. If you took mostly credit risk and were reliant on wholesale funding your very existence is threatened at the moment. To some extent the way I am shorting now is to find smaller banks that were consciously rejecting interest rate risk two years ago – but still have fat margins. They had to have done something bad to maintain those fat margins…

Margins and risk

I mention all this for a reason. If you are earning more than 300-350bps of spread in a bank you are taking some funding or credit risk – because it is simply not possible to earn that in interest rate risk unless you own some very exotic instruments.

Contra: I know of one bank which issues callable funding to buy callable assets. This sounds like it is matched – but if rates go up the funding gets called and they lose (having to refinance at higher rates). The assets however stick around. If rates go down the assets get called – and the funding sticks around and they lose. This bank has held up remarkably well through the crisis because they don’t take many credit risks – even if they do take enormous interest rate risks. The bank could still wind up highly problematic but it won’t get that way on credit.

But short of such exotic behaviours if you are earning more than about 300-350bps you are taking credit risks. If you earn 350-400bps of spread and you specifically disavow interest rate risk then you are taking a very large amount of credit risk. Find me regional banks that rejected most interest rate risk two years ago and I will find you regional banks that are potential credit problems.

Hybrid banks

Most banks fall somewhere in the middle. To pick one of many examples, Webster Financial Corp is a bank which is pretty well run. It is headquartered in Waterbury Connecticut and its management are known to several Connecticut hedge fund types.

Webster made the same mistake as me. It thought that interest rate risk had to be avoided. So it reduced interest rate risk – quite sharply. It decided to take some (quite a small amount) of credit risk. You would call this risk diversification. The stock price shows the result.

Webster was early to recognise their mistake. As I said – they are pretty well run– and in the scheme of things they are hardly a bank that is likely to wind up owned by the Feds. But shareholders have hardly had a nice time.

Webster took mostly interest rate risk. This is its five year margin summary – which shows non-performers including real-estate owned reach the princely sum of 1 percent of assets (on their way quite a bit higher).

Note the interest rate spread was always below 350bps but still above 300bps.

That is very high by global standards because in most countries banks do not have huge refinanceable assets – and hence do not carry the sorts of interest rate risks that American banks take. Its a middling spread by American regional bank standards.

My regional bank

Last week I referred to a regional bank which I had been working on with a reader. Small cap – so we won’t mention its name. I wanted to short the subordinated debt – but couldn’t get a borrow so I shorted the common. It took me about 15 minutes to make that decision.

Here are the things that stood out:

1. It had specifically disavowed interest rate risk saying that it had shifted its security holdings towards FHLB securities because they could easily be pledged (as they contain little credit risk) and could be purchased with short durations,
2. It had massively grown its deposit base in deposits greater than 100K by using promotional rates,
3. After this promotional binge it was the bank most dependent on jumbo deposits that I have ever seen. Given the noise of banks failing the populace could get quite jumpy about jumbo deposits and I would not think that funding base is very secure,
4. Despite having expensive funding and not taking interest rate risk it had a margin of 4% (now declining) - which suggests it was doing something else to get the margin,
5. That something was (more or less obviously) accepting credit risk. It had 1.75 times its shareholder equity in home-equity-line-of-credit products, 1.7 times in real estate construction loans and one times its capital in loans secured by vacant property owned by developers. It grew its HELOCs sharply in 2006. In 2006 it also more than doubled its mortgages on unimproved land.

I think there is a fair bet that this regional bank is cactus. I really wanted to short its debt.

But hey – what do I know? The provisions through the income accounts were less than 10 million – and outstanding provisions are about 20. The non-peforming loans are only half Webster Financial – and they think the lending is so good that they expanded loans in every one of these categories during the last quarter.

This of course leads to the real reason I won’t name the bank. I smell a rat. A big, nasty hairy one, but I can’t quite identify what it is. On this blog we don’t want to make it up and want to correct our mistakes – and we will not blurt out quasi-fraud allegations against small regional banks unless of course we have three decent forms of evidence.

The Wells Fargo puzzle

So now I will leave you with a question for which I do not know the answer. How is it possible that Wells Fargo’s margin 450bps? Please – serious answers are gratefully accepted as I do not understand.

This is not unusual in the history of Wells Fargo. There have been several points where their margin was greater than the prime rate.

Wells Fargo’s margin is off-the-scale high by global standards – I once did a global survey and it was the fattest margin major bank in the world.

Prima-facie that is good. High margins allow you to take considerable losses and remain profitable.

But usually have to take some risk to get high margins – and as the calculations above show – it is unlikely to be all interest rate risk. When I did my survey the other super-fat margin financials were subprime credit originators (mostly autos and credit cards).

As I said – I do not fully really understand just why Wells Fargo is quite so profitable. Can someone help?

Please...

Things I stuffed up – edition one - Interest rate risk versus credit risk

Anybody that trades stocks makes mistakes. I have made plenty. I would prefer sweep those under the carpet but a little bit of healthy self-flagellation is good for the spirit. Besides I hope it will make me a better investor. Besides I just posted that I purchased Ambac - something that could (easily) wind up as the next mistake. So you should know just how much I stuff up.

So this is the first of (almost certainly) many posts detailing things I stuffed up.

The list for the first choice is long. How about these?

(a) Believing that regional banks of Credit Agricole (which are very good) would offset the losses at the investment bank (which is very bad). Stock is down from 36 to 12.

(b) Believing that the mortgage insurers would blow up this cycle but the bond insurers would probably be OK. Ambac is down 90 to 1ish and is no longer writing much business. MTG (which was my favourite short) is down from 60 to 6 but is writing plenty of business. Got the wrong shorts… and didn’t short the bond insurers…

(c) Believing that the (seemingly extreme) valuation difference between News Corp and other media stocks would solve itself by New Corp’s stock price rising. It didn’t as a stock price comparison of Viacom, Time Warner and News Corp will attest. (It was a wash – all the stocks lost a little.)

(d) Buying Origin Energy at under $2 and selling it at about $4 on the basis that the utility parts of the business were fully recognised. I sold it despite loving the management. It is currently under hostile takeover at $15.60 – and the Aussie dollar in which it is priced has almost doubled. I didn’t recognise just how good the gas assets were. This was non-trivial as the fund I worked for owned almost 5% of the company – and left more half a billion dollars on the table and it was my fault.

Against this it should be pretty hard to tell what the worst intellectual error I made in the past five years is. But I have a candidate. I thought that the interest rate risk in US banks would blow up before the credit risk.

Background

The US has a very unusual mortgage market. Most mortgages have the peculiar term of being fixed rate when rates are rising – but being refinanceable if rates fall. This means that customers pay more for their mortgages than most jurisdictions – but that all the interest rate risks fall on the financial sector.

For instance in most markets the difference between central bank fund rate and the average mortgage rate is less (often much less) than 200bps. In the US it is unusual to get a conventional mortgage at under 6 percent – and the feds fund rate is 200bps. Mortgage margins in the US are more than double most countries.

For this however the system as a whole takes an awful lot of interest rate risk. If short rates were to go to say 8 percent there would be 5-7 trillion in mortgages that yield less than that. Individual institutions might say they were hedged – but the system as a whole cannot be hedged.

I spent an awful lot of time looking for banks and other institutions that were particularly levered to interest rate risk. WestAmerica Bancorp (an otherwise pristine bank) stood out. If you look at the balance sheet I linked in my previous post you will see that it contains $1.5 billion in fixed rate securities financed floating. That number is very significant compared to pre-tax income of 120 million or tangible book value of about 270 million. And WABC is by no means the largest offender.

My back of the envelope calculation was that the system had about 400 billion of pre-tax profits. That included all brokers, all banks, all insurance companies, fund managers – the works.

The US system had 7 trillion of interest rate miss-match. Almost half the profits of the entire US financial system could disappear in a 200bps rise in rates across the yield curve. And they would have disappeared without a penny of credit losses. A lot of institutions would lose their profits entirely. They would in my view all try to hedge simultaneously guaranteeing the dynamic hedging strategies that were in place did not work.

And I thought with Alan Greenspan setting the tone of the Fed the stuff up on inflation and hence interest rates was inevitable. Greenspan never saw a problem he could not fix by pumping more liquidity into the system. I thought Helicopter Ben was even more likely to use a little inflation to get the US out of its mess. Indeed that is where the “helicopter” moniker comes from – a speech to that effect. So essentially whilst I thought that credit problems were sort of inevitable – the US would inflate their way out – and hence the real manifestation would be an interest-rate-risk debacle.

So I spent a couple of years getting completely obsessed about interest rate risk. It led to some OK shorts (eg Fannie and Freddie) but meant I underestimated the credit story.

The credit risk I thought had been passed pretty heavily to the non-bank sector. It existed in the Europeans (I sort of knew about UBS). It existed in the investment banks (including Citigroup). It existed in some regional banks (I knew about Bank United). But I was stunned it wound up quite so bad at Fifth Third. Just stunned.

I thus covered a Fifth Third short many years ago. (Ooops.) I was short a bunch of interest rate risk sensitive banks (such as North Fork which was purchased by Capital One) and I didn’t short MBIA and Ambac. Indeed I was tempted to go long (but fortunately I did not). I made money on a few interest rate shorts – but altogether it was not a profitable activity.

A few years ago the short end of the yield curve was at about 1%. The long end in the 4s and quasi-government guaranteed mortgages were in the high 5s. Borrowing short to buy Fannie Mae backed mortgages was the seeming no-lose trade. Everyone was on it. It didn’t even carry much credit risk because everyone knew the government backed Fannie.

However it carried – and still carries – massive interest rate risk. Everyone seemed to ignore that. My usual reaction – if everyone is doing something then it will probably lose you money. I would rather be on the other side.

Still I remain convinced that this is a theme that will play out. Warren Buffett says inflation is heating up – and he doesn’t stretch the duration of his assets.

There are good people who think inflation is highly unlikely. Paul Krugman (who I admire) suggests that Bernanke should ignore the inflation naysayers. Mish writes for ever on how inflation is not likely – see here and here for examples.

I will get back to this shortcoming one day soon.

John