My Blog

October 5, 2007

I still can’t respond directly to comments because Sampasite is blocked in China, and the proxy I use doesn’t allow me to see and post the required access codes, so I have to post responses in the form of new entries.

JKH, you ask if the way a purchase were funded might change my conclusion that a Chinese purchase of a foreign bank would, if large enough, actually destroy market value.  You agree that this is the case for an equity-funded purchase but wonder what would happen if the purchase were funded by issuing debt.

The way I see it is that it would still cause a reduction in market value, probably even greater.  Let us assume that Bank A (a Chinese bank) buys Bank B (a foreign bank) and pays for it by issuing $100 of stocks.  We both agree that in this case there would be a reduction in total market value.

Now let us assume that the new (larger) Bank A issues $100 of debt and uses the proceeds to repurchase stock.  I think you would agree that the resulting capital structure would be the same as if Bank A had funded the purchase of Bank B by issuing debt.  The question is will this second transaction create or destroy value.

As I see it, since the new Bank A is almost certainly going to have more debt than is optimal, the “old fashioned” M&M framework makes it clear that the marginal cost (financial distress costs) of new debt will be much greater than the marginal benefit, so enterprise value will fall.  That should reduce the total value of the combined venture even further.  I realize that I am not answering your question directly but rather am backing into an answer, but my first reaction tells me that financing the purchase with debt would be even worse for total market value than financing it with equity because we are presumably on the wrong side of the marginal-value-of-new-debt curve.  Would you agree?

As for your second point, I think you are saying that the more ownership the government sells, the more it is willing to force the bank to enter into transactions that protect the creditors (which encompasses the interests of regulators) and harm shareholders.  I hadn’t really thought of that but if I am interpreting you correctly, then I agree fully.

Twofish, the idea that a share price is some sort of barometer of the health of a company is very widespread but, as you point out, wrong.  To be technical, this is only the case if the share price has a great deal of intrinsic value and little time value (the case for a very solvent, healthy company).  If it is all time value, then it reflects changes in volatility more than changes in “health”.  This, in a way, is the main point of my Far Eastern Economic Review article – Chinese bankers and their regulators should not misinterpret what the market is saying about Chinese banks.

I think the option framework does support your conclusion that banks need to be heavily regulated because the temptation (especially with deposit insurance) to speculate wildly is too great.  In a nutshell I believe that this is the story of the US S&L crisis of the 1970s-80s.  Once the banks were made insolvent by DIDMCA, they asked for significant relaxation in their lending restrictions.  Congress, unwilling to foot the bill for cleaning up the S&Ls, bought the argument that with more freedom the S&Ls could “earn” their way out of bankruptcy, but of course they were wrong.  The incentive for the insolvent S&Ls was to borrow more (deposit-insured) money and speculate wildly in the hopes of success.  Heads, I win; tails, the government loses.  Who wouldn’t speculate under such conditions? 

Not at all surprisingly, the S&Ls were the chief piggy banks for the then-exploding junk bond market and when heads turned up on the flipped coin, their investors made fortunes.  When tails turned up, however, which it did more often than not, the government took it on the nose.  As we all know, in the end the S&L clean-up turned out to be far more expensive for the US government than it originally would have been.  This becomes breathtakingly obvious when you use the option framework to analyze the incentive structure and predict the banks’ behavior.  In fact the option framework does an extremely good job of predicting a lot of otherwise inexplicable behavior.

Twofish, this is probably why I am so much more pessimistic than you are about the Chinese banking system.  The way I see it, the incentive structure for excessive risk-taking is too great.

By the way, if you don’t have an FEER subscription (get one – it has become a very interesting read again) you can read the original FEER article here: http://www.iea.usp.br/iea/english/articles/pettischinasvolatility.pdf

Comments (2) for "Responses to "Should Chinese...

Michael – Thank you for your response. We are in basic agreement. I agree an acquisition will result in reduced equity time value as a result of lower asset volatility in the combined firm. Your reasoning also makes intuitive sense that a further debt for equity recapitalization may actually reduce value. However, using the option framework, I continue to get a slightly different conclusion regarding the effect of recapitalization. This may be secondary to your main theme, but it’s an interesting aspect of the option framework. I’ll construct a simple numerical example, extending your example of a $ 100 stock buy back after the acquisition. Suppose the newly combined bank has $ 1000 in assets with $ 800 in liabilities. Suppose the market value of equity is $ 400. Then the intrinsic value of equity is $ 200 and the time value of equity is $ 200. Suppose the firm undertakes a stock buy back of $ 100, funded by $ 100 of new debt. The firm now has $ 1000 in assets and $ 900 in liabilities. With recapitalization, the equity call option is restruck at $ 900. The intrinsic value is now $ 100. Assuming the asset value and volatility haven’t changed, the probability distribution of upside values contributing to time value hasn’t changed. Upside risk on $ 100 of intrinsic value is the same as upside risk on $ 200 of intrinsic value. But intrinsic value has declined, so there is less offsetting downside risk. Therefore, time value has increased. Suppose the time value of the equity call option increases by from $ 200 to $ 250. There is an equal change of $ 50 in the time value of the put option on assets that is associated with the liabilities. The put option on assets has been restruck from $ 800 to $ 900 (the new value of liabilities). Assuming assets remain valued at $ 1000, the put option remains out of the money. With the new and higher strike, put option pricing now captures an additional $ 100 of downside volatility of asset value (nearer the money) – the same $ 100 of downside volatility that has been removed from the value of the equity call option. Because the put remains out of the money, its value is exclusively time value. This (absolute) value has increased by the same amount as the change in the time value of the equity call. Because one is long and the other short, the net effect on the time value of the entire capital structure is 0. This seems to make intuitive sense, because I’ve not assumed any change in the volatility of assets after the new bank is formed. Time value is an asymmetric decomposition of roughly symmetric volatility, and all asset volatility must be transmitted through the capital structure. Nothing can be lost. So recapitalization transfers time value from debt to equity. (Of course, it also transfers intrinsic value from equity to debt, if one extends the definition of intrinsic value to negative values.) As a separate issue, the pricing of the new debt would determine its market value as well as any changes in the market value of other liabilities outstanding. Debt is essentially the combination of a risk free asset, a short put, and a long interest receivable. With respect to new debt issued, if the present value of the interest receivable equals the change in value of the total put (i.e. if priced at marginal cost), the value of other liabilities outstanding should not be affected. It seems that the buyback has increased the time value of equity. It is in this sense I suggested it could make up for some of the equity time value lost in the bank acquisition. The resulting equity position would be riskier as an option. It would have a higher gamma value (accelerating option value accretion for increases in asset value) and a higher vega value (more value accretion for increases in the implied volatility of asset value). In short, and obviously, it is more highly leveraged as a result of the debt recapitalization. This aggressive capital management strategy would appeal to higher risk investors. It wouldn’t be very appealing to the government for the reasons you have outlined more generally. On balance it might not be very prudent. But this judgment also depends on whether the going in capital condition of the acquired firm includes sufficient equity to justify such in change in capital structure. . By jkh - 10/6/2007 11:10 PM

One reason I'm optimistic about banks is that when trying to look at how to fix them, the example of the US Savings and Loan crisis was brought up a number of times. As far as the big banks go, the banking regulators have been I think keeping a close watch on things, and part of the reason for bringing in foreign investors so that there would be another pair of eyes looking at things. It bears repeating that the big-four banks got into trouble in the first place, not because they were undertaking risky investments, but because they were ordered to issue loans based on criteria that would almost guarantee that they would go bad. They were ordered to do this because there was no other means of providing social welfare spending. After 1998, the big four banks were given a different set of orders. I'm much more worried about the joint-stock commercial banks because regulation there has been much looser. By Twofish - 10/7/2007 2:08 PM