Incentive Pay, Cigars and Petrus

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

Incentive pay is hard.

It feels like we don't get a lot of happy news these days, but here is some: Apparently Alphabet Inc. (formerly Google) paid the people working on its driverless-car project so much money that they all called in rich and left.

Early staffers had an unusual compensation system that awarded supersized payouts based on the project's value. By late 2015, the numbers were so big that several veteran members didn't need the job security anymore, making them more open to other opportunities, according to people familiar with the situation. Two people called it "F-you money."

The obvious lesson here is, don't pay people so much that they don't need to work any more. (I mean, that's the lesson for Alphabet! The obvious lesson for engineers is: Build a time machine, travel back a few years, and go work for Google's driverless-car project.) But there are other lessons. For one thing, lots of companies pay people so much that they don't need to work any more, and those people keep working anyway. They are typically called "chief executive officers," or "hedge fund managers," though sometimes they have other titles. The trick is not so much that you can't pay people tens of millions of dollars; it's that if you are going to pay people tens of millions of dollars, they have to be the sort of people who cannot be satisfied by any amount of money. (This might be because they don't care about money, though that is rare; more often, they are driven by competitiveness rather than money; in a few cases, they are just people who really, really like money.) It pains me to write this, but this may explain why no one has ever paid me tens of millions of dollars.

It is not a pure quantity problem, though. The compensation system was designed to give employees equity or quasi-equity in the self-driving car business (now called Waymo), rather than tying their compensation to the broader company. But of course the business ... doesn't exist. It's not public, there's no stock price, and it's never sold a car. So the quasi-equity was "based on periodic valuations of the division" and agreed operational milestones:

Part of the problem was that payouts snowballed after key milestones were reached, even though the ultimate goal of the project -- fully autonomous vehicles provided to the public through commercial services -- remained years away. 

Incentive pay for a salesperson at a big profitable public company is relatively straightforward: You see how much money she brought in, and pay her some portion of that. You have the basic quantity problem -- pay her too little and she'll leave in a huff, pay her too much and she'll leave in a yacht -- but you have a pretty good sense of what she's worth, and of how much money is lying around to give her.

It's much harder to design incentive pay for an engineer at an early-stage project that might be fantastically valuable but that won't make money for years. Deep down, you don't want to pay her for engineering accomplishments or operational milestones: You want to pay her a ton of money when the division turns out to actually be profitable, and nothing until then. You want her to take the risk of failure. She is unlikely to want that deal. (Presumably she's an engineer at Google, rather than a startup founder, in part out of risk aversion.) But if you pay her up front, you take the risk of the project not working out. And if you pay her too much up front, that risk can be self-fulfilling: All the engineers might work until they complete enough milestones to get rich, and then quit before you ever sell a car.

Lobbying as a business model.

Here is a sad, amazing story about Evan Morris, a pharmaceutical lobbyist who committed suicide after being investigated for allegedly embezzling from his employer. As a non-expert, I frankly found it a bit hard to tell what was embezzlement and what was lobbying. So for instance Morris "smoked cigars and shared a bottle of Petrus" with Virginia Governor Terry McAuliffe at a fundraiser: That sounds like lobbying? He could probably expense the Petrus? But also: "Prosecutors are investigating whether Mr. Morris took company money to pay for real estate, golf memberships, Rolex watches, fancy wine and cigars." I'm sure he did spend company money on golf and wine! But golfing and wining seems to have been his job.

Other times, though, it seemed really really obvious to me what was embezzlement and what was lobbying. But, again, I am not an expert, and the experts seem to have found it less cut-and-dried: 

In one example, Mr. Morris hired Mr. Courtovich’s Sphere Consulting in 2012 for $880,000 to do policy work with think tanks, according to documents viewed by the Journal. Genentech paid Sphere two payments of $440,000 each on Nov. 1 and Dec. 1.

On Dec. 10, Mr. Courtovich’s firm sent a payment of $448,986.22 to Mr. Morris’s personal bank account.

Eric Lewis, a lawyer for Mr. Courtovich and Sphere Consulting, said the payment was to reimburse Mr. Morris for personal funds that he said he used for an event with the American Enterprise Institute, a Washington think tank. Mr. Lewis provided the Journal with an AEI invoice for $448,986.22 that Mr. Morris gave Sphere.

An AEI spokeswoman said the invoice was falsified.

Wait, what? Morris hired a consultant to do work for his employer (Genentech Inc., owned by Roche Holding AG). Nine days after Genentech's check cleared, the consultant sent half of it back to Morris's personal account. I know what that looks like! But, no, the consultant was sure that it was for a legitimate business expense. Because that is apparently how lobbyists work? They pay for expensive events out of their own pockets and then are reimbursed, not by their employers, but by consultants hired by their employers? Anyway, in this particular case, the business expense was fake, no doubt to the deep surprise and chagrin of the consultant. Also in another case:

In another instance, Genentech paid National Media $2 million for public affairs counseling and strategic consulting, the records show. On April 6, 2012, days after National Media received the final installment of $750,000, the firm sent $303,048.95 to the Hacker Boat Company, which made and sold the Mulligan, Mr. Morris’s boat.

Evan Tracey, a vice president at National Media, said Mr. Morris had submitted an invoice for that amount saying it was to reimburse Mr. Morris for renting space and paying for food and other expenses related to an event at the Hacker Boat Company for the “Democratic attorney general group.”

It wasn't, but again, I am sure that the consultant was just flabbergasted to learn that the boat it bought for Morris days after getting paid by his company was not a legitimate business expense. The lobbying business is amazing, I tell you what.

Is Marxism better than index funds?

Here's a fun Medium post from Matt Bruenig about how to achieve socialism by (1) taxing wealth in various ways and (2) using the money raised to start a giant "social wealth fund" by which the government will own the means of production. You might think that this idea would be appealing mainly to socialists, and you'd be right, but there is also a sly appeal to people who are worried about index funds:

Creating a robust system of actively-traded social wealth funds might actually save the capital markets from the rise of passive investing and the similar problems associated with that rise. As Inigo Fraser-Jenkins noted last year to much fanfare, management of equity has increasingly moved from active traders that try to determine where the market is going in order to make investment decisions and towards passive funds that just buy and hold the whole market (or big swaths of the market) indiscriminately.

We talked last week about the depredations of index funds and "quasi-indexers," who -- according to various theories -- reduce competition, lower investment, and generally impede the efficient allocation of capital. It all sounds like a critique of socialism. And yet index funds seem so clearly to be the highest expression of financial capitalism, the perfection of science and efficiency in finance. It's like the telos of financial capitalism turns out to be socialism. When we talked last year about the Fraser-Jenkins note that compared index funds -- unfavorably -- to Marxism, I three-quarters-jokingly speculated that the ultimate benefit of indexing and algorithmic investing would be to solve the socialist calculation problem and allow an efficient collectivized ownership of the means of production. "The market is the best algorithm ever developed for allocating capital," I wrote. "So far! But it also creates incentives for someone to build a better algorithm." Maybe it will be socialism.


Here is a funny little Securities and Exchange Commission enforcement action against Sidoti & Company LLC. Sidoti is "an independent research firm, primarily focused on small and microcap public companies," but over time it also expanded into investment banking and, briefly, running a hedge fund. The problem is that it was a pretty small firm and basically one guy did all of those things at once:

During this period, Sidoti’s CEO controlled Sidoti’s investment banking and research departments and maintained trading authority in the Fund. Although Sidoti had written policies preventing the misuse of MNPI by its investment banking and research departments, nothing in Sidoti’s written polices prevented Sidoti’s CEO and its associated persons from misusing MNPI obtained from these departments when making trading decisions for the Fund.

If you get material non-public information about a company by being its investment banker, you can't then give it to your in-house hedge fund to trade on it, or to your research department to write client notes about it. And various vice versas. Sidoti didn't have sufficient procedures to prevent that sharing. It had a restricted list of stocks it wasn't supposed to trade, but that only made things worse: "Between November 3, 2014 and May 5, 2015, however, there were 126 instances when the Fund traded in a stock that appeared on the Daily Restricted List." Sidoti settled for $100,000.

We talked about Sidoti in June 2015, when they were in the news as the delightful research firm that only gives Buy and Neutral recommendations. The point of this institutionalized optimism is to get companies to "market with us," as a Sidoti executive wrote, setting up meetings between company management and Sidoti's investor clients. (This is the point of a lot of sell-side research.) Management meetings were the cornerstone of Sidoti's research business -- "our accounts value management visits more than anything else that we do" -- and you can see how that would lead to confusion. If you think that research is done by meeting with company managers, and that investing is done by meeting with company managers, and that investment banking is done by meeting with company managers, then it wouldn't seem that critical to keep all those meetings separate. Company managers tell you their plans, and you act on them, by doing their banking business or writing research reports about them or buying their stock or all three. Refusing to trade the stock because you met the managers with your investment banker hat on just seems fussy.

Trolling as a business model.

We talked last week about how Andrew Left's Citron Research seems to have developed a strategy of shorting a company's stock and then trying to goad Donald Trump into tweeting mean things about it. It's a good strategy, but as I pointed out, Left's execution -- 13-page research report, relatively restrained tweet -- left something to be desired. Here is Sahm Adrangi's Kerrisdale Capital with a similar strategy but a bit more gusto:

Yes, I like it, well done. Now be sure to include @realDonaldTrump when you tweet the actual report. Also buy ads on Morning Joe. "He gives us so much material to work with," Adrangi told me by e-mail. "Also, frankly he's been great for short activism. Market at new all-time frothy highs, random companies getting undeserved stock surges, etc."

Market structure.

Here are Giovanni Cespa and Xavier Vives on "High frequency trading and fragility":

We show that limited dealer participation in the market, coupled with an informational friction resulting from high frequency trading, can induce demand for liquidity to be upward sloping and strategic complementarities in traders’ liquidity consumption decisions: traders demand more liquidity when the market becomes less liquid, which in turn makes the market more illiquid, fostering the initial demand hike. This can generate market instability, where an initial dearth of liquidity degenerates into a liquidity rout (as in a flash crash). 

It's a pure market-microstructure-economics model, by the way. Nothing here turns on fat fingers or glitchy algorithms or spoofing; nothing turns on computers at all. It's just that individually rational behavior by traders sometimes leads to crashes, and that computers allow those traders to be even more individually rational (and faster). As I once wrote about flash crashes, "the explanation is not some weird exogenous accidental move to computerization and algorithms, not some failure of regulation or imagination." It's that the algorithms are good at being selfish and rational, and sometimes that leads to crashes.

Rich people having fun.

"Steve loves parties," says Larry Gagosian about Steve Schwarzman, who threw a 70th birthday party on Saturday:

“It was brilliantly stimulating,” said Koch, the day after Schwarzman’s party. “You learned a lot about Asian theater. There were acrobats, Mongolian soldiers and two camels. It was a little bit of everything.”

That's a little bit of a few highly specific things, honestly, but it's not like I go to a lot of parties with camels. I suppose that once I start going to Park Slope children's birthday parties that will change.

Elsewhere, Eike Batista is no longer a negative billionaire, though he is in prison. But at least he got to fling some money into the ocean:

Among the allegations in the proceedings is that Batista, acting on advice from a spiritual adviser named Ubirajara Pinheiro, tossed about $130,000 worth of gold coins into the Atlantic last year from the deck of a yacht festooned with flowers and perfumes for the occasion.

“All those riches that everyone talked about," Pinheiro said in a phone interview from Rio, "I don’t think that brought him good fluids.” So, Pinheiro explained, he advised Batista to make amends with the sea goddess Iemanja by giving gold back to nature after his years of mineral extraction.

That is pretty dumb but it could be worse. The gold probably just sank. Batista has an oil company, too; imagine if Pinheiro had told him to dump oil in the ocean to appease Iemanja.

People are worried about unicorns.

Ben Thompson worries about Snap Inc., the Disappearing Unicorn, in a fascinating post:

To summarize, Snap’s strategy is to:

  • Deliver innovative and differentiated products that…
  • Cost a lot to deliver but…
  • Capture the best customers…and PROFIT!

That’s definitely not Twitter; indeed, the real analogy for Snap is from another part of technology entirely: it’s Apple.


Not only is Snap not promising a traditional moat, it is in fact selling its humanity as a company. That the company and its Steve Jobs-admiring CEO in fact do understand users better than everyone else, that that will result in a sustainable differentiation, and that the prize will be the top end of the advertising market.

This could be true? My problem with the analogy is that I cannot for all my efforts understand Snapchat, just on a product level. What does it do? Why? These are rarely questions that I have about Apple products, except the Apple Watch. But this is probably my problem, not Snap's. It does not feel good to know that "the top end of the advertising market" has left me behind. Elsewhere: "The invention of the toilet was 'much more important than Facebook,' says economist Joseph Stiglitz." And: Dedrone.

People are worried about bond market liquidity.

"The Coming Drought in Corporate Bonds" is about reduced issuance (due to tax-law changes), but it's a liquid-based metaphor so I'm going to include it here.

Things happen.

Credit Suisse Plans to Cut Up to 6,500 Jobs This Year After Loss. RBS Said to Plan More Than $1 Billion of Expense Reductions. Banks Eyeing Dublin After Brexit Face Trader Shortage. Toshiba's Chaotic Earnings Raises Doubts Over Grip on Business. Toshiba chairman to resign amid $6.3bn writedown. Banks Look to Cellphones to Replace A.T.M. Cards. Co-operative Bank Puts Itself Up For Sale. Abu Dhabi’s New $125 Billion Sovereign Fund Is Taking Shape. Commerce Nominee Wilbur Ross Will Keep His Stake in Chinese-Government-Backed Company. Former Chief of A.I.G. Tries to Fight the Headlines. Trump risk factor. Fierce storms ahead in 2017 ‘shareholder spring.’ Did Dodd-Frank really hurt the US economy? Kyriakos Chousakos and Gary Gorton: "Post-crisis regulatory changes have aimed at restoring bank health, but measuring bank health by Tobin's Q, we find that the ill health of banks in the recent U.S. financial crisis and the Euro crisis has persisted, especially compared to other crises in advanced economies." The Resolution of Distressed Financial Conglomerates. "The current framework in antitrust—specifically its pegging competition to 'consumer welfare,' defined as short-term price effects—is unequipped to capture the architecture of market power in the modern economy." Legal Insider Trading Profits Often Amount to Peanuts. Playboy is featuring naked women again. Disney Severs Ties With YouTube Star PewDiePie After Anti-Semitic Posts. Cheap, Very Last-Minute Valentine’s Day Gifts You Can Buy at Duane Reade.

If you'd like to get Money Stuff in handy e-mail form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at

To contact the editor responsible for this story:
James Greiff at