The spread between the return on 10-year Treasuries and Baa-rated corporate bonds was 336 basis points last week. In historical terms, that is very high. In fact, the only times in the last 50 years that it was higher were the trough of the tech wreck and the 1982 recession/stagflation busting cycle.

High credit spreads make it harder for corporations to borrow and invest. As spreads widen, the economy tends to slow.

For investors, however, high spreads represent additional potential returns for a given unit of risk-taking. Peak levels of risk premia typically precede strong returns on risky assets - even if the strong returns are short-lived, as was the case in 2003.

Spreads have hooked down slightly as the credit woes that peaked following the Bear Stearns (BSC) debacle may finally be stabilizing. Time will tell whether that was “the” peak or whether the current easing is simply a lull.

Regular readers know I follow the spread between Baa-rated bonds and Treasuries to get a feel for how much they are willing to accept risk.

Last week, as the market rallied on the initial news of the JP Morgan (JPM - Annual Report) and Fed-led rescue of Bear Stearns (BSC) I said “The big questions are whether, how much, and how quickly the spreads will begin to narrow again. That will be the real signal from bondholders that the worst is behind us from the current crisis.”

Instead, credit spreads continued to widen. They averaged 343 basis points last week, up from 340 in the prior week.

It remains a double-edged sword. As long as the spread continues to widen, risky assets will perform poorly. However, the abnormally high risk premia we are currently seeing indicate that longer-term investors will be paid more handsomely for accepting such risks than they have been paid on average in the past.

By all accounts, JPMorgan (JPM - Annual Report) has all the leverage to complete its proposed acquisition of Bear Stearns (BSC) for $2.00 per share. Actually, since it is a stock deal and JPMorgan shares have risen since the announcement, the acquisition price is now $2.40 per share.

Still, for a deal likely to be made at $2.40 per share, why would anyone be willing to pay $5.00 per share for the stock today? One answer that keeps getting bandied about is that bondholders are buying the stock in order to vote for the deal, as it is worth losing some money on the stock to preserve the value of the bonds.

That argument makes some sense, but I believe it doesn’t tell the whole story. It is possible for the bondholders to not only preserve the value of their bonds, but to lose very little on the stock in the process.

While the stock has been getting all the news, options on Bear Stearns have seen enormous trading volume. I think the bondholders are buying the stock, but hedging their bets by creating synthetic short positions.

A synthetic short consists of writing a call option and buying a put. Here’s how I think a bondholder can be playing this:

Buy 100 shares for $5
Write an October $5 call option for 2.10 (bid price as I write this)

Buy an October $5 put for $2.45 (ask price as I write this)

Net cost for this transaction is $5.00 + 2.45 - 2.10 = $5.35. At expiration, it will be worth $5.00 no matter where the stock is trading. So, for $0.35 per share, or $35 for every $500 of exposure, the bondholders can buy the right to vote on the deal.

Given that the bonds were trading at $700 per $1,000 face value on Friday, and are worth $1,000 when backed by JPMorgan, it is pretty simple math. For every $1,000 of bond exposure, you can pay $70 to vote in favor of a deal that is worth $300 to you.

What’s more, since there was so much more debt than equity, only a small fraction of the bondholders need to make this bet to gain an overwhelming majority of the equity votes.  Or, each bondholder could insure a smaller portion of their value.

Taking that into consideration, paying $5.00 for the stock starts to make sense.

Provided, that is, you already own the bonds.

How quickly value can evaporate these days. Bear Stearns (BSC) closed on Friday at $30, down from $57 the night before. I spent the weekend working up an article on why investors should still avoid it despite an apparently low price relative to book value. Before I could publish it (not that it would have done any good), JPMorgan says it will buy Bear Stearns.

JPMorgan Chase & Co (JPM - Annual Report) said on Sunday it would buy stricken rival Bear Stearns for just $2 a share in an all-stock deal valuing the fifth largest investment bank at about $236 million.

How could a stock go from $57 to $2 in two trading days, all the while having a book value above $80? I can’t say, exactly. But here’s what I was planning to say on the assumption that the stock would still be $30:

Many value investors look for stocks trading at prices below the company’s book value per share. Such stocks are often undervalued.

After Bear Stearns’ shellacking on Friday, the stock closed at $30. Yet on the conference call, Bear Stearns CEO said its book value “fundamentally” is still in the mid-$80 range. Does that mean value investors should step in?

I don’t think so. Taking a look at the 10K the company filed in January, I found it quite easy to make most of that $11.8 billion book value disappear.

To start with, there is $950 million recorded as “goodwill & intangible assets.” Subtracting this out leaves a tangible book value of $10.8 billion, or $82 per share.

Next, Bear’s balance sheet includes, in “other assets,” $5.2 billion in “financial instruments that are valued using models or other valuation methodologies.” Most of the total is “comprised of financial instruments whose fair value is estimated based on internally developed models or methodologies utilizing significant inputs that are generally less readily observable.”

Given that the credit issues have really just started to flow through the system, I’d want to be conservative about the value of these instruments. For the sake of argument, I’m assuming they are really only worth half their book value. Admittedly, I’m just making a guess. But then again, so is Bear Stearns.

If I take $2.6 billion off the value of these instruments, the tangible book value is reduced to $8.2 billion, or $60.22 per share.

Bear also reported $33.5 billion in “assets of variable interest entities and mortgage loan special purpose entities.” These were the loans it was unable to move off of its balance sheet. Of this amount, Bear says its maximum loss is nearly $3 billion. For conservatism, I’ll assume that maximum loss gets realized. That cuts the tangible book value to $38.

Finally, As of November 30, 2007, the Company had notional/contract amounts of approximately $13.40 trillion of derivative financial instruments, of which $1.85 trillion and $1.25 trillion, respectively, were listed futures and option contracts. These amounts are not fully reflected on the balance sheet.

It’s likely that many of these derivative instruments offset each other, reducing the total exposure Bear faces. But without being able to see the underlying contracts, investors are unable to make that judgment. Since the tangible book value is less than a tenth of one percent of the notional value of these contracts, it isn’t hard to imagine the contracts changing in value sufficiently to wipe out the remaining tangible book value.

While there may be a buyout of Bear Stearns, it would be made by sophisticated investors with full access to the company’s financial information. These investors are in a position to determine a value for Bear Stearns.

I, on the other hand, am not. And if I can’t make a reasonable estimate of the company’s value, I’m going to stay away.

Disclosures: William Trent has no positions in the companies mentioned