Kinder Morgan is de-MLP-ing.

When I left for vacation last weekend the talk of the town was all about tax inversions, where American companies decide that they'd prefer not to pay U.S. corporate income taxes and just stop. But the tide seems to have changed in my absence: First Walgreen decided not to invert, and then yesterday energy pipeline giant Kinder Morgan announced that it would roll up its master limited partnerships -- non-tax-paying, non-corporate entities that pay out much of their income to shareholders1 -- into its corporate-tax-paying, American, parent. Here's DealBook:

The deal exposes one of the significant challenges M.L.P.s face: the need to constantly acquire new assets to continue increasing the dividends they pay investors.

Smaller M.L.P.s continually buy new assets to satiate this need. But Kinder Morgan’s related companies had grown so large that it was difficult to find suitable targets. Furthermore, because the companies were paying out such large dividends, they were hard pressed to comfortably finance big transactions.

The vague sense in U.S. corporate tax law is that if you're a thing that retains earnings then you should probably pay taxes on those earnings. But if you're a thing that just sprays all your money out at your shareholders -- an MLP, or a real estate investment trust, say -- then you don't have to pay taxes. Not paying taxes is a pretty attractive proposition, but if you run a business being bigger and more acquisitive might be even more attractive.

Anat Admati is big on bank capital.

This weekend's New York Times profile of Anat Admati doesn't mention my favorite thing about her book, which is that it quotes liberally from Microcosmographia Academica, but I guess why would it. It's mostly about her advocacy of much higher bank capital requirements: "I don’t mean 20 percent. I don’t mean 30 percent. I mean add a digit. I mean a lot more." There's a lot of vague blather about speed-limit metaphors but don't let that fool you, this topic is deep and fascinating and a good test of how to think about a bank. Like, where to start in estimating how much capital banks should have? You might say something like "well, JPMorgan has $2.5 trillion of assets, and 95 percent of the time it will lose no more than $49 million in a day, so its $160 billion of common equity tier 1 capital -- almost nine straight years of one-day-in-20 losses -- seems more than adequate." If you say that you fail the test, of course, though it's interesting to think about why. Or you might, as John Cochrane does, point to "the Modigliani-Miller point that the cost of equity declines if banks issue more of it," but of course nobody believes in Modigliani-Miller so it's an uphill fight to convince anyone that this is true. Or you might consider the demand for safe assets, and point out that the product that banks produce -- the main thing that society wants from them -- is safe-looking debt. If you restrict the amount of safe-looking debt that they can produce (by requiring them to replace like a quarter of it with risky-looking equity), then it seems sort of self-evident that there will be some negative consequences, though perhaps safe-looking debt is not a thing that society should be trusted with.

Blackstone is trying to automate its investment bankers.

I mean, sort of:

Its latest investment is in Ipreo, a capital markets data provider that it bought with Goldman Sachs from KKR in April. Ipreo can support Blackstone’s brokerage unit, its growing capital markets operations and its GSO credit arm by making it easier for Blackstone to deal directly with those who buy its debt and equity and that of its portfolio companies, instead of paying Wall Street to act as its agent.

The best thing here is that Ipreo "is now working on applications for investors that predict the deals they may be interested in based on past buying patterns." It used to be, your local independent investment banker would get to know you personally, and would put you into deals based on detailed knowledge of your tastes and preferences. Now it's just an impersonal recommendation engine saying "Customers Who Looked At This Bond Also Bought ...," and what has been lost in that transition?

Don't sign a five-year non-compete.

Doesn't that sound way too long? Christopher Rokos left Brevan Howard in 2012 after earning some $900 million there trading interest rates, and for some reason he now wants to work again. The reason might have to do with London real estate, which I hear is wildly expensive, particularly given Rokos's "168-page proposal to triple the size of his Notting Hill mansion using a four-storey basement extension to include a swimming pool, cinema, climbing wall, and gym." Someone should really do a profile of an English hedge fund manager who hasn't submitted elaborate planning documents for architectural follies, if any exist. Anyway Rokos wants to start his own hedge fund and is bummed that he can't until 2017, arguing that the long non-compete period "would 'atrophy' his trading skills, and damage his professional reputation and contacts." Honestly that seems fair enough, though I guess the question is, why did he sign it in the first place? (And quit in the second?) You could have a whole Wall Street career in five years, and who knows what trading -- or Rokos's skills -- will look like in 2017?

Don't issue capital appreciation bonds.

I'd be the first to admit that I am not an expert on municipal finance, but my extremely non-expert rule of thumb is, if a corporation wouldn't do a thing, a state shouldn't do it either. Here are Cezary Podkul and Felix Salmon on goofy structured tobacco-settlement bonds that capitalize all their interest for ages, so for instance Ohio "got $319 million up front, and promised to pay a whopping $6.6 billion at maturity." That sounds ... like the wrong thing to do? But it's not clear that it actually was. The tobacco settlement won't actually generate that $6.6 billion, and the bonds are non-recourse, so they'll default. Basically this means that Ohio will never see another cent from its tobacco settlements, long past the maturity of the bonds. But that means ... that it sold its tobacco-settlement revenues for more than their present value? Seems fine actually.

Things happen.

Here is a rousing defense of M&A bankers by an M&A banker. People still don't like Wall Street for some reason. Calpers may cut back on hedge funds and active management. One upside of giant international financial scandals is that different countries' financial regulators are getting better at working together. The SEC won a pump-and-dump jury trial. Alibaba's strategy involves ... graymailing? ... luxury brands. The Argentina situation is pretty much where I left it. Izzy Kaminska's decline and fall of the Roman Empire. Operation Character Flaw.

1 I know, unitholders.

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.