On Monday, as a bevy of banks were settling a zillion dollars of mortgage lawsuits and putting themselves on a path to (1) certainty and (2) giving money back to shareholders, Goldman released a research note with the results of a survey of investors' expectations of bank capital return.1 Here is what some sample of investors expect:

Total payouts are expected to increase to an average of 58% post-CCAR/CapPR from 43% in 2012. ... The survey results suggest the biggest increases in dividend payout ratios will be for Citi and Capital One, while PNC and Morgan Stanley are unlikely to meaningfully move higher. For buybacks, investors expect the biggest increase for BB&T and JP Morgan (vs. their actual buyback, not vs. 2012 approval levels), while there is little change expected for Morgan Stanley, Bank of New York and Northern Trust. ... Many of the banks with the most variability of responses are those that are coming off subdued capital deployment levels in 2012, including Capital One, Bank of America, Citigroup and Regions. Given the lack of consensus, it seems that regardless of the announcement, the market is likely to be “surprised”.

I too prefer to order my life so that I'm surprised by everything.2

Anyway the interesting/disappointing part for me is what investors thought about what GS calls the "Mulligan rule." This refers to the fact that, in the 2012 bank stress tests, banks asked regulators for approval to return an amount of capital, and if the regulators said no then the banks basically couldn't do anything (ex regular dividends etc.) for another year, but in the 2013 tests if the regulators say no the banks can go back and ask once more for another, lower amount of capital return. I was pretty bullish on this: the do-over gives you a chance to be more aggressive once, and scale back if regulators say no, so you'd think that at least some banks would be aggressive and get away with it, while others will be too aggressive and have to cut back to a more moderate capital return but still no harm no foul. Or so I would think. I am in the minority:

And here, conveniently, is why banks wouldn't be aggressive - because their own shareholders would get mad at them for being too aggressive:

Isn't this disappointing? If a bank asks to return $10X of capital to shareholders, gets rejected by regulators, and compromises at $8X, there are two ways of looking at it:

• Management did its job - of trying to maximize returns to shareholders - but was thwarted by regulators doing their job, of, y'know, regulating that, which is how things are supposed to be; or
• Management's job was predicting, and getting along with, its regulators, and getting this wrong and being forced to take the slightly humiliating make-up test is an indication that management is not good at that job.

The first theory is kind of how, like, everything else works: regulators propose prudential rules that require banks to hold lots of capital, reduce risky activities, etc., and bank managements quietly but zealously lobby to pare back those rules to allow them to take more risks on behalf of shareholders. It's an adversarial system: the banks probably want too much risk-taking, the regulators probably want too little, and they meet in the middle. As, like, a world, you might not be too keen this structure - it might depend on whether or not you think that a fight between banks and regulators would be a fair fight - but as shareholders, why should you care? Shouldn't you want management working for you, and trust regulators to shut off anything too egregious?

But the second theory is what shareholders seem to actually think, at least about the stress tests. If you're a bank it may be smart to lobby behind the scenes to loosen regulations and free up more capital to return to shareholders. But it's dumb to look overly aggressive in front of your regulator. The business you're in isn't maximizing returns to shareholders as much as it is optimizing regulatory relationships. If you get the second one right, the first will follow.

1.Goldman Sachs Equity Research, "Investor expectations high for 2013 CCAR/CapPR results, particularly for dividends," January 7, 2013. Sorry no link.

2.But then I am not an efficient market. Nor, I guess, are the 50 investors GS surveyed. One neat thing is that the GS researchers look at option implied dividends and finds that they are 17% below survey expectations, particularly on Citi where options imply a 3% dividend payout ratio and the survey expects 11%. What does this mean? Did GS survey a particularly optimistic group of investors? Are option implied dividends only sort of a thing? Or, when presented with a survey asking "how much will [Bank] pay out," are investors in [Bank] likely to interpret the question as "how much would you like [Bank] to pay out?"? Since after all [Bank] would hate to disappoint investor expectations. Etc.