Goldman in Tech and a Crash in China

Also Yanis Varoufakis's payment plan, a Finra violation, and the usual buybacks-and-liquidity.

Goldman Sachs, tech company (1).

Here is the story of how Goldman Sachs has gotten more involved in investing in, advising and partnering with tech companies. There is some awkwardness ("From the looks of the crowd, they didn't want to hear EDM," says the DJ at Goldman's SXSW party, and Lloyd Blankfein says that at Goldman they "try to disrupt ourselves"), and the requisite stuff about hoodies and culture clash, but what struck me is how much of the Goldman-as-venture-capitalist-and-tech-partner pitch just looks like a regular old investment banking pitch:

Entrepreneurs say Goldman's pitch in some ways mimics Andreessen Horowitz's, offering small companies a range of resources. "They have a whole menu of choices," says Larry Augustin, CEO of SugarCRM, an enterprise software company backed by Goldman's private capital investing group. He says the firm has introduced him to a team dedicated to helping startups with everything from basic administrative troubleshooting to broad strategic insights. For example, Augustin says, Goldman helped him benchmark his customer acquisition costs against those of his competitors'. Another entrepreneur says Goldman provided assistance crunching numbers and assessing potential acquisitions. This kind of expertise, two guys in a garage could never re-create.

Also, they're good at fundraising. (Disclosure: I used to work at Goldman.) I suppose if your startup model is two guys in a garage plus a kooky billionaire writing checks, this level of service is impressive, but of course public companies use bankers for "assistance crunching numbers and assessing potential acquisitions" all the time. The innovation here is bringing public-company-level investment banking to private companies, which makes sense if the private companies are all raising hundreds of millions of dollars at billion-dollar valuations. 

Obviously part of the goal is for Goldman to win roles advising on big initial public offerings, but if you get enough of a private-company ecosystem then maybe the IPO becomes less important. This is revealing:

"In this world, the men and women who run the company are the ones who decide who they want as investors," says Jerry Yang, a co-founder of Yahoo! who now backs startups. "The rest of us, as investors, have to earn those rights to be a part of a round."

Once you go public that dynamic flips. So why go public?


Tom DeMark, the guy who became famous for putting a copy of the 1928-1929 U.S. stock market chart on top of a 2013-2014 U.S. stock market chart, has now put a copy of the 1928-1929 U.S. stock market chart on top of the Shanghai Composite Index chart. That is ... what you do when you really like that 1929 stock chart? I imagine him walking around all day with a transparency of that 1929 chart, holding it up to everyone he meets and squinting to see if it fits. 

The thing about the 1929 chart is that it always ends in a big crash, so when you overlay it on everything else in sight, you end up predicting big crashes and talking like this:

"The die has been cast," said DeMark, 68, the founder of DeMark Analytics in Scottsdale, Arizona, who has spent more than 40 years developing indicators to identify market turning points. "You just cannot manipulate the market. Fundamentals dictate markets."

He made similar statements in February 2014 about the Standard & Poor's 500 Index, saying that if certain conditions were met, U.S. stocks had reached a point resembling the time before the 1929 market crash. The S&P 500 rallied 8 percent over the next two months. He said Monday that those conditions didn't materialize at the time.

Uh huh. It was nonsense at the time, but it's even more nonsense in China, because one its chart doesn't even look like the 1929 chart, but also two because China's market is, you know, structurally different from the U.S., insofar as its market is run by a Communist Party and all. In other news, a spokesman for the China Securities Regulatory Commission "said the regulator had received complaints about the sell-off on Monday and that it would take enforcement action soon against those who engaged in what he said was 'coordinated stock dumping,'" and "Western banks say they are coming under heavy pressure from Chinese officials to refrain from negative comments."

Goldman Sachs, tech company (2).

Yesterday the Financial Industry Regulatory Authority fined Goldman Sachs $1.8 million for "systemic Order Audit Trail System (OATS) reporting violations spanning a period of more than eight years, failure to accurately submit required trade reports to the appropriate FINRA Trade Reporting Facility (TRF), and related supervisory failures." Goldman's dark pool generated a lot of reportable order events -- like, customer submits order, customer cancels order, that sort of thing -- and Goldman was required to report data about those events to Finra. This is fiddly data and Goldman messed it up once or twice or 63 billion times:

During the Second Review Period, the Firm transmitted to OATS approximately 42,189,531,270 inaccurate and/or incomplete required ROE data elements. Specifically, the Firm: (a) submitted Order Cancellation Flags incorrectly indicating that orders cancelled by the Firm were cancelled by a customer, or indicating that orders cancelled by a customer were cancelled by the Firm between January 1, 2013 and January 3, 2014 (approximately 28.2 billion instances); (b) submitted inaccurate Order Origination Codes indicating that orders were received from a customer or originated from within the Firm instead of indicating the orders were received from another broker-dealer between October 17, 2011 through February 20, 2015 (approximately 11.3 billion instances); 

It goes on. The first review period involved 6.3 billion inaccuracies, the third 15.1 billion, and the fourth just 38.7 million inaccuracies, which really seem hardly worth mentioning on top of the other 63 billion violations. Goldman is paying $1.8 million, or less than 3/1000ths of a cent per violation, which probably represents some sort of bulk discount.

One assumes that this is not nefarious, nor all that important, but on the other hand having a consolidated audit trail of orders is sort of a key part of regulatory oversight of electronic trading, and learning that the audit trail was wrong for eight years is ... troubling. Like, bad on Goldman -- it's a good thing that Goldman is friends with a lot of big-data startups, maybe one of them can help -- but also bad on Finra, no, for not noticing for eight years?


Here is Yanis Varoufakis defending his plan to hack into his own ministry's tax records to set up an alternate payments system:

While I understand the press's excitement emanating from elements of that exchange, such as having to consider unorthodox means of gaining access to my own ministry's systems, there is only one matter of significance from a public interest perspective. There is a hideous restriction of national sovereignty imposed by the troika of lenders upon Greek ministers who are denied access to departments of their ministries pivotal in implementing innovative policies. When sovereignty loss, due to unsustainable official debt, yields suboptimal policies in already stressed nations, one knows that there is something rotten in the euro's kingdom.

Here's a transcript of Varoufakis's now-famous call. Guan Yang argues that Varoufakis could have achieved the same result by launching an open data initiative: Release some taxpayer data publicly (enough to identify everyone), and then use that data to set up the new payments system. Here is a New Yorker profile of Varoufakis describing his tenure as a "five-month-long TED talk," and if that is true then I guess I should watch more TED talks. And here is a review of a book that "is as near to a McKinsey report on classical Greece as has ever been written."

Goldman Sachs, tech company (3).

I don't really get why "Manhattan District Attorney Cyrus Vance Jr. will challenge the dismissal of charges against Sergey Aleynikov" on appeal. Why is it a good look to pursue this poor guy -- who has been declared innocent by multiple judges -- to the ends of the earth at what kind of looks like the behest of Goldman Sachs? Is this prosecution politically popular somewhere? Or is there an important principle here? Will New York corporations not be safe from an epidemic of code-stealing if Aleynikov isn't sent back to prison? Or has Aleynikov just become a prosecutorial white whale, an object of pure obsession unmoored from rational considerations? Elsewhere, consider the "responsible corporate officer" doctrine, which (maybe) allows courts to send corporate officers to prison "for failing to prevent statutory offenses by subordinates, even if they themselves were not aware of any wrongdoing." People talk a lot about wanting to cut back on mass incarceration of nonviolent criminals, but they don't actually want to stop incarcerating any particular nonviolent criminals.

People are worried about stock buybacks.

Quick who said this:

In too many cases the real culprits of short-termism are the boards and the C.E.O.s that do buybacks which they know will promote their stocks in the short term and make their options more valuable.

If you answered Hillary Clinton or Larry Fink, good effort, but wrong. The surprising answer is Carl Icahn, who is better known for constantly pestering Apple to do more stock buybacks. But to every thing there is a season, etc. Icahn is quoted in Andrew Ross Sorkin's somewhat skeptical column on Hillary Clinton's "quarterly capitalism" proposals. Sorkin thinks that Clinton's ominous mention of stock buybacks might be "code for an argument by some — Senator Elizabeth Warren included — that stock buybacks could be considered stock manipulations," so the 2016 election may yet be fought over revisions to Rule 10b-18.

Elsewhere in quarterly capitalism, here is a Roosevelt Institute report, and accompanying blog post, about "Defining Financialization." E.g.:

Beyond the growth of the financial sector itself, the way the finance industry has come to dominate the priorities and thinking of the real economy as a whole is a major feature of financialization. A key aspect of this is the rise of the shareholder primacy model of corporate governance. The current belief that corporations exist solely to maximize shareholder value is the product of an ideological, legal, and institutional revolution that resulted, among other consequences, in the reallocation of corporate funds away from investment and toward payouts to stockholders. This has led to a situation in which the purpose of finance is less about getting money into productive enterprises and more about getting money out of them.

On the other hand, according to Credit Suisse strategist Michael Mauboussin (in this Barron's article): "The proclivity to return cash to shareholders hasn't changed over the decades, Mauboussin said, although the means have shifted to favor buybacks over dividends." But dividends aren't manipulative, you know.

Meanwhile, here is the story of Mylan's rejection of Teva's takeover advances. When shareholders complained, urging the company to take the deal, this happened:

Executive Chairman Robert Coury leaned across the table and retorted, in language laced with expletives, "This is a stakeholder company, not a shareholder company," according to multiple attendees, meaning his constituents went beyond investors and he wasn't obligated to agree to a tie-up.

Mylan is based in Pennsylvania, but it is a Dutch company since it inverted to the Netherlands this year, and "Dutch policy makers have spent the past decade touting the benefits of Dutch law to global corporations as part of an effort to turn the Netherlands into a management-friendly bastion." I suppose if you worry about quarterly capitalism you could always take some tips from the Dutch.

People are worried about bond market liquidity.

Jon Macaskill is pretty meh on liquidity worries, though he does say that "It is now a mark of the serious individual in finance to issue a dire warning about the threat posed by a lack of market liquidity." Izzy Kaminska worries about a cognitive deficit fueling a liquidity crisis, quoting UBS strategists who argue that "the growth of the credit markets have not been matched by the addition of research resources (e.g., credit analysts) in many of the silos," and worrying that "the true underlying factor driving bond liquidity risk could be a discrepancy between the sheer volume of available securities in the market and the number of sophisticated/expert brains capable of scrutinising them properly on behalf of investors who lack such skills."

Things happen.

Cardiff Garcia on the reverse repo facility. "Becoming a hot shot in finance may be the best way for a bright graduate to help the global poor." The Tom Hayes Libor trial has gone to the jury. Fannie Mae is getting a fancy new headquarters, just in case you had doubts about its future. Consultant Cliffwater criticizes Carlyle's Claren, concerning clients. "At a state-run amusement park in Argentina, the government is encouraging children to take aim at virtual vultures, a not-so-veiled reference to the nation’s international creditors." Bloomberg Billionaires calculates Donald Trump's net worth at $2.9 billion. Here is a story about Donald Trump's lawyer, among other topics. The $100 Billion Deal Is Still Out There for Pharma. "Given decent public policy, an automation-driven productivity surge is nothing to be afraid of." B-Schools Aren’t Bothering to Produce HR Experts. "We find that people who experienced greater intensity of the Earthquake become more risk tolerant." Aggregation is fine. Mexico Cilantro Fields Strewn With Feces Prompt Partial U.S. Ban. "Also this kills me to say but you should read the entry for Paleolithic Lifestyle." Dog staircase.

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 on this story:
    Matt Levine at

    To contact the editor on this story:
    Zara Kessler at

    Before it's here, it's on the Bloomberg Terminal.