Levine on Wall Street: An Oil Plunge and a Gold Stockpile
Oil is getting cheaper.
My most recent linkwrap was on Wednesday, when West Texas Intermediate crude was above $73 a barrel; it got below $64 today and, hey, here's a headline saying "Oil at $40 Possible as Market Transforms Caracas to Iran." So, big changes. Also: "Prepare for the long-term fall in energy prices," says Nick Butler, who argues that "Technology in many different ways is adding to supply and reducing demand" for fossil fuels. But the immediate catalyst for the recent drop is OPEC's decision not to cut production in the face of weaker demand and a glut of competition from U.S. shale and Russia. You can read estimates of oil production breakevens, at U.S. shale plays and elsewhere, here or here or here or here or here or here. Vladimir Putin is pretty chill about things, despite "basing next year's budget on oil at $100 a barrel." Should governments hedge oil price risk? Unsurprisingly, the oil price slide is bad for oil company stocks. But also for bonds: The U.S. high-yield market is increasingly correlated with oil, and Josh Brown worries that that could be bad for stocks, since "consequence-free" high-yield liquidity has fueled buyback activity in stocks.
The Swiss won't be stockpiling gold.
"The 'Save Our Swiss Gold' proposal stipulating the Swiss National Bank hold at least 20 percent of its 520-billion-franc ($538 billion) balance sheet in gold and never sell any bullion was voted down by 77 percent to 23 percent, the government said yesterday," and just that sentence gives you a pretty good sense of what a dumb idea it was. (Willem Buiter: "Forbidding a central bank from ever selling any gold it owns reduces the value of those gold holdings to zero," though I guess if you got really desperate you could have another referendum to reverse the first one.) Now the Swiss National Bank will have more room to defend its cap on the Swiss franc's value versus the euro, and also won't have to bring back tons of gold that it stores in Canada and the U.K. At times like this you might as well check in with Zero Hedge, who are holding up stoically ("an unprecedented scare campaign by the Swiss National Bank"). Elsewhere in Swiss referenda, "Swiss voters refused to give up a 152-year-old tax break for rich foreigners in Geneva and other wealthy areas that the government says helps the economy," and as an American I have a hard time thinking about a referendum asking people "do you want the government to stockpile gold and cut tax breaks for rich foreigners?" Surely in the U.S. that would be a double yes, right?
If you like market structure you'll like this Federal Reserve discussion paper about dealer networks in the municipal bond market. The Fed's researchers looked at trading patterns and found that some dealers were at the core of the market -- dealing with lots of other dealers and customers -- while others were at the periphery. "Central dealers are more efficient at matching buyers and sellers than peripheral dealers, which shortens intermediation chains and speeds up trading," so if you want immediacy you should go to a central dealer. They're also better at providing liquidity during times of market stress, and they do less pre-arranged riskless-principal trading and more holding of bonds in inventory. But there's a trade-off: The central dealers charge more, about twice as much as peripheral dealers in this muni bond sample. Also fun is how markups are split when there are multiple dealers between the seller and buyer of bonds:
We find that the dealer closest to the ultimate buyer earns the largest share of the overall profits, irrespective of how many dealers are involved. This pattern reflects the common perception in the industry that bonds are not bought, they are sold. The municipal bond market is known to be a buyer’s market (Schultz, 2012), and identifying investors willing to buy is considered to be dealers’ most crucial and costly task.
One story of high-frequency trading is that making markets in stocks is not an activity in which humans have any vast advantages over computers, and computers are much faster and cheaper than humans. So small high-frequency trading firms disrupted the human-driven business at big banks and exchanges, and now markets in stocks are pretty much made by computers. Markets in bonds are still sometimes made by humans because they do have some advantages; see the item immediately above this one. What about pricing and approving consumer and small-business loans? I don't know, it's not hard to believe that those might be better done by computer. LendingClub and Prosper think so, and they are driving big banks to get into the computerized lending business too.
"Investment staff at the $298.9 billion California Public Employees' Retirement System, Sacramento, received $8.7 million in bonuses in the year ended June 30," up 14 percent from the previous year. The irresistible comparison is to Pimco, which is about six times as big as Calpers, but which paid its managing directors $1.5 billion in 2013, or around 170 times as much as Calpers. You'd have to say that Calpers employees look like a bargain, though of course they invest a lot of Calpers's money with outside asset managers (including $1 billion with Pimco) so it's sort of fees on fees. But very small fees on fees, as these things go.
Some Goldman-Libya hijinks.
There's a long-running dispute between Goldman Sachs and the Libyan Investment Authority over a bunch of option trades that Goldman sold Libya that went horribly wrong. Here's a story from that dispute that I can't quite make sense of. The gist of it is that Goldman sold the LIA a call option on Citigroup stock, but instead of calling it a "call option" -- which everyone knows can lose all its value if the stock finishes below the strike price -- Goldman documented it as a forward plus a put, which is economically identical but, I guess, sounds less scary? I have my biases, but I find it hard to believe that that was the reason; usually there's some reason of accounting or optics or something behind a structure like this, not just "the customer won't figure out that he could lose all his money." Though I could believe that the structure was designed to take advantage of other possible misunderstandings. (The LIA says that it "did not properly understand whether the Disputed Trades involved direct equity investments, or a species of quasi-share ownership, or constituted an entirely synthetic financial instrument.") On the other hand, Goldman also provided the LIA with a "scenario analysis" showing how much the LIA would make if the Citi stock ended up at various prices, all of which were well above the strike price of the call options. Which is not a particularly useful scenario analysis, and is I guess support for the theory that Goldman was hoping that LIA wouldn't notice that it could lose money.
Chris Hohn has to pay his ex-wife $531 million in their divorce. Fund management is pretty sexist. Debt collector buys for-profit colleges, in a good illustration of the Coase theorem. "The highest court in New York, for the first time, affirmed a century-old principle that banks’ offshore branches are distinct entities when enforcing asset seizures and beyond the reach of creditors for the purposes of judgment enforcement." Izzy Kaminska on copper bullion. George Packer on Angela Merkel. Ronald Barusch on C&J Energy and Nabors. John Dizard on Ukrainian booby-trap bonds. The Federal Reserve on the demand for short-term safe assets. JW Mason on monetary policy ("it is central planning that cannot speak its name"). George Pearkes on the correlation between volatility and returns. Black Friday Google trends. Crisis-era bonuses are vesting soon. The Laguna Beach high school reunion.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Matt Levine at email@example.com
To contact the editor on this story:
Zara Kessler at firstname.lastname@example.org