Skip to main content

Goldman Research Analysts Suggest That Morgan Stanley Should Get Smaller And/Or Better At Bond Trading

  • Author:
  • Updated:

Wall Street banks' research on their competitors is not only a window into analysts' anxieties about their own banks' prospects, but also a ripe area for conflicts between investment advice and industry advocacy. The days of analysts writing research reports that were like "Facebook should really do a huge equity offering and hire my bank as sole underwriter," or whatever, are mostly behind us, but when banks write about their industry you might wonder if they're giving dispassionate advice or pushing their employer's interests. And when they write about their competitors it must be tempting to be a bit underminey. So various banks have published research saying "actually breaking up the big banks would be bad for shareholders," which may be true but also not un-self-interested, as breaking up the big banks would surely be bad for the equity research analysts they employ. And then Morgan Stanley published a don't-break-up-the-banks piece saying "... except Citi,"1 and if you knew that MS is in the process of trying to buy a chunk of Citi cheaply you might be like hmmm. Today Goldman recommends that Morgan Stanley get out of fixed-income trading, and, again: suspicious!

That's arguably not their main point; much like JPMorgan last week, GS set out to quantify how much of their fun financial regulation is ruining, which again you could read as advocacy. Even though it's the opposite of what Lloyd is advocating. Here's Bloomberg:

New bank regulations and capital requirements are “structural” changes to the industry that are more to blame for declining profits than the U.S. economic slump, Goldman Sachs Group Inc. analysts said.

“The operating environment is unlikely to change any time soon, and we see shareholders of challenged banks becoming more demanding in asking management teams to lay out a path to unlocking value in the near term,” analysts led by Richard Ramsden in New York wrote in a report published today.

Their view contrasts with Goldman Sachs Chief Executive Officer Lloyd C. Blankfein, who said in November, “I don’t think we can conclude that the slowdown is secular rather than cyclical change.”

Here is their main chart, which is sad though perhaps too soon to call secular:

So what's the evidence for the regulatory explanation? Various things, including this chart, which I quite liked:

Basically if

  • you want a mortgage in a post-Basel-III-ish world, and
  • your bank is a buy-and-hold mortgage lender, and
  • banks are attempting to replicate the return on equity that pre-Basel-III-ish mortgages brought them, and
  • let's just say other assumptions go here,2 then
  • you have to either put down 40% or more of the purchase price, or pay a lot more for your mortgage.

Most relevantly, if you want to put down only 5%, and you probably do, then according to Math your mortgage should have a 9.39% interest rate. If it did Bill Gross would be so happy! Anyway, though, if you believe that, then post-crisis changes in capital requirements will either (1) encourage prudent, low-leverage mortgage lending and/or (2) make housing totally unaffordable for most people. Regulation, you double-edged etc.

But my most favorite part of Goldman's note was the part where they tell Morgan Stanley to get out of FICC trading. Almost:

While its current situation is undoubtedly frustrating, we argue that MS could use its stock valuation to its advantage by pulling back from fixed income trading and using capital freed thereby to repurchase its own shares, increasing book value and returns. We estimate a 50% RWA reduction in fixed income would free up a marginal $12bn in capital. If this excess capital were used to repurchase shares at current market price, ROTCE would increase nearly 200bp and tangible book value 25%.

Even with the resulting loss in revenue from this pullback, this could still be a positive for shares. We estimate that if a 50% reduction in FICC RWA resulted in a 70% reduction in fixed income revenue, ROTCE would remain flat. That said, this non-linear relationship implies that if MS can insulate the impact to revenue by shifting the fixed income business towards more flow-driven and capital-efficient assets, a large reduction in risk-weighted assets could be meaningfully ROTCE accretive.

The biggest risk attached to a dramatic fixed income risk-weighted asset reduction is a corresponding decline in trading revenue. That said, MS’s plan for a 25% reduction in RWA by the end of 2013 is only about half the relative size of what several European investment banks are targeting. While still relatively early in the de-leveraging process, EU banks with active risk-weighted asset reduction plans (e.g., UBS, Credit Suisse, Royal Bank of Scotland) have been performing largely in line with those that have not (e.g., Deutsche Bank, Barclays, BNP).

So - that last part seems to actually be true; here is a chart:

Basically, banks that have announced significant cuts to fixed-income trading have not really lost ground in fixed-income trading to banks that are not making those cuts. That somewhat undermines the regulation-is-killing-us thesis, doesn't it?

Does it? One imaginable view here is that new regulations, rather than just spoiling the fun for banks, has forced them to be smarter and less risk-tolerant in running their businesses without really forcing those businesses to scale back. Another I suppose would be that banks always win and that any regulatory costs fall on market participants, not banks: so, for instance, lower bank inventories might mean less liquidity for markets, but not lower trading profits for banks. "Regulation is killing us" for an "us" that is not just the banks.

Or maybe things are just complicated. I like the contrast between Goldman's very precise a priori mathing out of how much a 5%-down mortgage should cost now - 9.39%! - and its vague "gosh there seems to be no correlation between cutting FICC balance sheets and losing FICC revenues." The theoretical math is easy and precise and totally wrong.3 The empirical results of banks' actual reactions to increased regulation and capital requirements are quite a bit messier.

Goldman Sachs Analysts Decide U.S. Bank Slowdown Isn’t Temporary [Bloomberg]

1. I mean, not really, but kind of. It's an August 29 note called "Hypothetical Bank Break-Up Would Destroy Value" and here is the relevant chart:

So breaking up Citi would create value! They walk this back in the text but there that chart is. INCIDENTALLY how do you like new-style footnotes? This is your chance to mourn for *** and friends.

2. Like, I guess, mainly that the only determinant of credit quality and capital is LTV, which seems counterintuitive, but y'know they're simplifying it's fine.

3. That link will change so, for posterity, the current rate for a 5%-down mortgage appears to be 4.25%. I'm not sure I believe that either, but it's closer than 9.39%. Incidentally by "wrong" I mean "not an accurate predictor of reality" - I'm not saying Goldman claims it is; they don't say "these are mortgage rates" but rather "these are the rates you'd need to charge to get back to where you were, if you wanted to do that, and if your assumptions were ours."