We've heard a lot about government bailouts in the past few months. So what exactly are these bailouts? After hearing some more griping about one of them - the part of the Troubled Assets Relief Program (TARP) that was devoted to stable banks like Wells Fargo and Bank of America - I decided to read up on some of the details.
It appears that this $250 Billion program works as follows. The federal government purchases preferred stock from the banks for cash. The banks then pay the federal government interest at an annual rate of 5% for the first three years, and 9% thereafter. That's a pretty high interest rate when the discount rate - the rate at which banks can borrow from the Federal Reserve - is running at less than 1%, but if a bank is in trouble, the money might help them survive a run by depositors. The banks are allowed to pay the money back, but not until the first three years are up.
Most of the funds weren't used to help banks that were in trouble though. Rather, stable banks were pressured into the program. Assuming that this was not just a cynical way for the government to extract above market interest rates from successful banks, what were Bernanke and treasury department trying to do?
Well, one clue comes from an exchange that happened during that meeting was when one of the bank presidents asked if the government "investment" was to be treated as debt on their balance sheet. According to normal accounting rules it is: it's money that has to be paid back eventually, and on which interest is paid in the meantime.
Bernanke and Paulson, however, agreed it should be treated as equity, and that they would change the rules if necessary to do so.
Why is this important? It's important because there are rules governing banks that require them to maintain debt to equity ratios within certain limits. Adding money to the "equity" side of the balance sheet allows the banks to pile up more debt - that is, lend more money. Adding money to the debt side of the balance sheet would do the opposite: it would constrain lending further. The government wanted the banks to lend more, not less, to avoid an economic downturn, so the money was categorized as "equity".
Of course, calling it equity doesn't make it so; the banks still have to pay interest on it, and they still have to pay it back after three years. Some investment banks might have lent to the limits of the regulations anyway, but those banks are, not coincidentally, bankrupt now. The banks that got the first round of TARP funds were some of the stablest and most conservative banks in the country - and those banks are likely to act as if the funds are debt, even if their balance sheets list them as equity. They're not going to use the money to extend even more loans that put them closer to going under.
What's the net result? The big, stable banks are burdened with interest payments to the government that restrict their lending just when it's thought the economy needs easier money. Plus they get tagged with the "bailed out" label when the reality is closer to the banks propping up the government.
Emergency Economic Stabilization Act / TARP
Wow, that's the first that I heard about interest payments on the TARP money. I had thought that I had heard on several occasions that the reason that stable banks were forced to take the money was to make sure that everyone would take money. In that if only unstable banks were taking money, that it might make some so-so bank decide to not take money and then fall into bankruptcy because they might be labeled as 'unstable' if they decided to take the money.
Looks like I have some reading to do.
Yeah, the nature of the so-called "investment" comes as a vast surprise to everyone who hasn't noticed it yet. And I find it pretty depressing how much even the financial press continues to get the details, as far as I can tell, very wrong.
I realized what was going on after I read State Street's 3rd quarter 10-Q filing, under the heading "Capital Purchase Program". Here's the summary as I understand it: (And I assume that the other big banks that were in the initial $125 billion had the same provisions; not sure about everyone who came after that)
State Street issued $2 billion worth of preferred stock, which was bought by the Treasury. This Preferred stock has no voting rights (typical of preferred stock, but not universally true), but in the event of a liquidation, preferred shareholders get money before common shareholders get it. (That's what Preferred means). There is a set liquidation value on each share; i.e. each share was sold to the Treasury for $100K and can be redeemed by State Street for $100K, under certain conditions. And there are the interest rates as noted above. Also, the company is prohibited from raising its dividend or doing a stock buyback while the Treasury owns preferred shares. (More of a comedy item at the moment, since we just cut our dividend to a penny a share...) In addition to this, the Treasury has options to buy 5+ million shares of State Street stock at over $50 a share, which isn't too interesting right now. Those 5+ million shares would be a little more than 1% of State Street's outstanding shares. I've read some news talking about the possibility of the Treasury converting its preferred shares to common stock, but as far as I know, the shares aren't convertible. (Though, as it turns out, the Treasury can invite the CEO to a meeting and then refuse to let him leave the room until he's agreed to something, so, YMMV.)
One may wonder how this Preferred Stock is different from a loan or a bond (i.e., as noted above, *debt*). The answer is, not very--there are some technical differences, but this particular preferred stock appears to me to be much more like a bond than like common stock. Unlike a bond, it will never come due, so in that sense it's a better deal for the company than a bond. But then, there's that spike in the interest rate, which is not a feature of most bonds, even in this environment.
One note about the discount rate and the prime rate--they're for short-term loans only, for dealing with normal liquidity requirements, not for long-term debt. I think a direct comparison of the CPP dividend rates vs. any of the Fed rates is misleading. My understanding is that the best direct comparison for those interest rates would be to the bank's corporate bond yields, which vary depending on market conditions and the bank's bond ratings. I don't have enough KS: Bonds to tell you what bond yields for new issues from the big banks are right now, but less than a month before the Treasury bought $125 billion in preferred shares, Goldman Sachs sold $5 billion of preferred shares to Berkshire Hathaway with a dividend of 10%. A brief survey of news items on recent bond issues and preferred stock deals shows a lot of scary-high dividend/yield numbers. My sense at the moment, with admittedly poor data, is that 5% for mid-term money is actually a good deal if you need cash, which banks need pretty badly right now. Tangentially, all of the complaints I've heard from bank CEOs have been about the attached strings (realized and hypothetical), not the dividend rates.
I agree that I don't expect any bank, at this point, to treat that money as equity for purposes of strategic planning. I'm not sure it's quite the lead weight that you're seeing it as, though.
Goldman Sachs is an investment bank. Given how investment banks were dropping like flies last year, of course they paid a big premium. That's a whole different animal from commercial banks like Wells Fargo, which is much of my point.
The yield on Bank of America bonds maturing in three years is 2.02% (BAC.HGP is the symbol I see, matures Jun 2012), so that would be a good comparison with the 5% the government is getting. I stand by my position that the 5% rate is usurious for the stabler commercial banks - which is probably why Wells dared to ask if they could refuse it (and were basically told "no").
Edited at 2009-02-19 04:15 am (UTC)
I see that BAC.HGP is one of the issues that's backed by the Treasury, via the TLGP. I'm not sure the TLGP limits for a company can be derived from public information, but it looks to me like it's not generally going to cover a company's total debt burden, or come even close. Perhaps you can correct my understanding there.
Do you have any other recent BAC bond issues handy of, which are of significant size and aren't government-backed, which are yielding less than 5%? I see a lot of fiddly little issues, but mostly I'm getting the impression that they were all special-purpose issues or have provisions that's not obvious on a casual examination, which affect what yields people are willing to take.
Hey, it was hard enough for me to find a commercial bank bond issue at all, let alone TLGP coverage information; I was actually looking for Wells Fargo, and couldn't find anything. The likely reason there aren't a lot of bond issues is that commercial banks tend to take deposits rather than issue bonds. However, I can give you two other pieces of information.
First, 2% was the typical range for bond issues with 2012 maturity for stable corporations who were not banks. The issues I remember seeing were General Electric issues.
Second, we can look at 3 year deposit rates. For Wells Fargo, the three year IRA CD pays 2.66%; the regular three year CD pays 3.1%. While those are more than 2.02%, they are both still much cheaper than 5%.
Incidentally, Citibank only pays 1.25% on their 3 year CDs, which is less than the 2.3% they are advertising on their 12 month CDs. I don't know if that reflects a different interest rate projection than Wells has or what. Citibank is kind of weird, though, as they are the only bank with both substantial commercial banking and substantial investment banking activities.
CDs are FDIC-insured, so as far as comparisons go, I think they have the same fundamental problem as a TLGP-backed bond. GE sounds like a better comparison, but it is well-known for having primo bond ratings, whereas the banks generally started a notch or two below GE and most have been downgraded at least once.
I'm looking at listings on FINRA's web site and also at Fidelity's individual bond listings. (Fidelity's site doesn't have the whole selection as far as I can tell, but I was scraping for any source that seemed to have useful data on individual bonds.) I'm not thrilled with the searchability and presentation of either of those sources. Where are you looking? We're getting very different impressions of the data, so I'd like to see what you're seeing.
Both of my sources have some Wells Fargo issues:
CDs are, like deposits, only partially FDIC insured; remember that there's a limit that affects many accounts.
More importantly, the fact that they are insured does not change the fact that they are the most common way for banks to take on debt to finance loans - and thus that it's quite legitimate to compare them to other forms of debt and debt like instruments, including the TARP funds.
You could legitimately argue that the interest rate ought to be adjusted for the insurance premium that banks pay for FDIC coverage. Adding the 0.05 to 0.07% insurance premium doesn't change the equation significantly, though.
The bottom line is still that banks have many ways to raise money at substantially below the 5% TARP funding rate.
Since it's been over ten years since I was actually in the business, I don't have good data sources at my fingertips. bakedweasels would be infinitely more helpful. But if you poke around on FINRA's website a bit more, you can find this: corporate bond yield stuff which has a graph in the upper left of high grade corporate bond yields. It's a composite of issues of various maturities, so it's not precisely comparable, but it's in the ballpark. Right away you can see one important phenomenon, which is that yields have swung up and down by over 300 basis points over the past five months. A 5% interest rate in October 2008 (when TARP was revised to the preferred stock plan) probably looked pretty sweet. A few months later, not so much.
Corporate bond spreads over Treasury yields have been awfully high. They aren't terribly specific about what index they're referring to, but BusinessWeek cites spreads of 600 bp over Treasuries in December, down to 470 bp last Tuesday.
But even that's not totally equivalent. The order of priority of creditors is extraordinarily important here. Debtholders and all depositors have a claim before any preferred stockholders get a shot. For a company with virtually no chance of default, the difference is minor, but if the market is seriously concerned about the future of the issuer, the difference can loom pretty large.
As an example of preferred stock issued by a bank, check this out: A Look at Citi's Trust Preferreds. Citibank's trust preferred stock, which actually has precedence over other preferred stock, closed Friday at $6.11 which given its annual payout of $1.62, is a yield over 25%. And note that just on Friday, the price swung between $4.40 and $6.33, i.e., the yield varied from 25% to over 35% over a single day.
(Incidentally, if you read a few more articles on the Seeking Alpha website, you'll quickly know a hell of a lot more about this situation than I currently do. This is especially informative.)
With regard to the TARP preferred stock, we should know that most preferred stock already counts as Tier 2 capital for the purposes of the Basel capital adequacy standards, so it isn't appropriate to pretend that it's just like debt; it isn't the highest-quality Tier 1 capital, like ordinary equity, but it does count as capital rather than debt. The unusual thing about the TARP preferred equity is only that it's classified as Tier 1 rather than Tier 2, so it's counting for more equity than it ordinarily would, but they haven't allowed anyone to treat as equity anything that used to be treated as debt.
Ordinarily, even reclassifying your lower quality capital as high quality capital would be asking for trouble, allowing more leverage than is appropriate, but this is like a methadone clinic for leverage addicts. If they go cold turkey, they're going to start having seizures and drooling all over themselves; this is a way to gradually deleverage while not completely halting all lending.
When you are looking at real preferred stock and at longer term bonds, the appropriate TARP interest rate for comparison is the 9% rate that kicks in at 3 years, and not the short term 5% rate. Preferred stock counts sort of as capital because the principal need never be repaid; if the banks do something similar by hanging on to the TARP funds indefinitely, they're going to be paying 9% for most of that period.
Also, the other limitations that harrock mentions become much more important if the funds are held and the restrictions stick around indefinitely. Not being able to raise your dividend for three years isn't that big a deal; never ever being able to raise it is a very big deal.
this is like a methadone clinic for leverage addicts
Except that it isn't. The leverage addicts are the investment banks like Goldman-Sachs and certain credit corporations like GMAC, and they weren't included in the original TARP group. Citi is borderline - and TARP may be a wash for them - because they have an investment banking arm as well as a commercial banking arm. Stable commercial banks like Wells Fargo and Bank America are getting the short end of the stick.