As Lloyd Blankfein fights against whispers that he will step down as Goldman Sachs CEO later this year, questions about the firm’s strategy are increasing.
Goldman Sachs’ fourth quarter earnings call on 17 January was a chest-baring attempt to convince analysts it was about to turnaround its flagging FICC trading business. “Low volatility, low activity” was the official explanation for the lacklustre results, which saw institutional client revenues hit a record annual low of $11.9 billion, about half the amount earned in 2010.
That was before the burst of market volatility in early February, when the VIX index returned to levels not seen since 2011. Meanwhile, commodities, a mainstay of Goldman’s FICC business, are increasing in price, which may prompt more trading. Goldman’s first quarter earnings are still over five weeks away, but might a recovery be on the cards?
Risky Finance has gone through seven years of Goldman risk filings to explore what has changed in the bank’s trading business. How much potential is there for mean reversion (as volatility indicates) and how much change is due to secular trends that the bank couldn’t or shouldn’t try to reverse?
Firstly, we incorporated the trading histograms published by Goldman in its annual 10-K filings. These show the number of days where net trading revenues fell within various ranges. In the chart below, click on the year buttons to see how this frequency distribution changed over time since 2010.
The most striking aspect – noted by Bloomberg last week – was the dearth of trading days on which Goldman earned more than $100 million. Between 2010 and 2016, the firm averaged 40 such days per year. In 2017 it only achieved four of them.
To understand the significance of this we have prepared another chart. We multiplied the number of days in each revenue band by the midpoint of the range to estimate the days’ contribution to Goldman’s annual trading revenue. For the bands on the edge – the days of $100m plus gains or losses – we conservatively created a slightly wider band.
As we can see from the chart, the sum of estimated revenues from all the bands closely tracks Goldman’s actual institutional client revenues during this period (these are higher because of items such as securities services or commissions that aren’t part of the daily trading P&L).
And now we can clearly see what Goldman’s problem is. It isn’t a problem with trading profits per se – the revenues generated by trading days below $100 million haven’t changed that much. Goldman’s problem is the disappearance of those $100m plus days that used to generate the lions’ share of overall revenue.
So what is going on? Here are five reasons why those $100m plus days might not be coming back, even if volatility spikes up again.
This is the juiciest reason. Recall how Goldman once made $600 million on a single day on the back of Greece’s debt-concealing swaps in 2001. The bank has a history of huge, profitable transactions – including Abacus, Libya, 1MDB and others – that it probably shouldn’t have done.
True, Goldman emerged relatively lightly from the swathe of misconduct fines in the wake of the financial crisis, and escaped from foreign exchange rigging allegations completely. But the reputational impact of controversies like Greece has taken its toll. Add to that the burden of new transparency regulations such as MiFiD II (which requires block trades to be disclosed), and one of Goldman’s traditional pathways to $100m plus days is closed off.
One reason that Goldman was able to make such outsized trading profits in the past was that the bank’s pre-Dodd-Frank capital framework vastly underweighted the risks in its trading portfolio. From trading book securitisations (such as the Abacus CDO) to single security exposures not captured in a value-at-risk model or swap counterparty risk, regulatory capital charges came thick and fast from 2012 onwards.
The Federal Reserve began to zero in on common equity tier 1 ratios in its Dodd-Frank stress tests, which forced the banks to reduce risk-weighted assets in order to retain their capital freedom. As the Risky Finance banking visualisation tool shows, Goldman’s reduction in market RWAs since 2012 has been dramatic.
And if you measure Goldman’s trading performance against regulatory capital consumption, there’s actually good news. From earning a 77 per cent return on market risk and counterparty capital in 2011, Goldman boosted that return to 93 per cent in 2017.
One reason that Goldman enjoyed 122 days with $100 million plus revenues in 2010 and 2011 was that it was then still allowed to do proprietary trading. The Volcker Rule stopped the bank from using its market making business as a hedge fund.
Absent a repeal of the Dodd-Frank Act, there is no chance of Volcker going away. However the good news for Goldman is that the Fed’s new head of banking supervision, Randal Quarles, is talking about easing the Volcker rule, in particular the way it applies to market-making inventory.
Combined with a bit of market volatility, that might boost Goldman’s ability to achieve a few more $100 million plus days.
Fixed income has long been a bastion against efficiencies that transformed other financial markets. Even as new regulations sought to reduce the lock that large banks had on over-the-counter trading of bonds and derivatives, clients still picked up the phone to a dealer rather than clicking a mouse.
In 2017, that bastion began to finally crumble, enough to merit a comment in Goldman’s 10-K filing: “the increasing volume of trades executed electronically, through the internet and through alternative trading systems, has increased the pressure on trading commissions”, the bank said, acknowledging that a collapse in bid-offer spreads had hurt its market-making business.
It remains to be seen how the new trading platforms will hold up in the event of significant fixed income turbulence, where Goldman’s traditional ability to offer liquidity might be more compelling that in is now.
Commodities have been a mainstay of Goldman’s FICC operation, to the extent that the bank developed significant infrastructure to be able to hold physical commodity assets, offsetting or hedging its positions in derivatives. That enabled Goldman to book handsome profits as a market maker.
Then, US Congress’s subcommittee on investigations exposed the depth of Goldman and other banks’ activity in controversial commodity trades. Other banks such as Morgan Stanley saw the writing on the wall and quit the market but Goldman stayed in.
Then in September 2016, the Fed took full aim at Goldman, proposing a rule in which Fed-regulated banks would be banned from physical commodity investments. In February 2017, Goldman was the only large bank to object against the proposed rule in the Fed’s public consultation.
Since then, the Fed has been silent, and it may be that Quarles and Fed chairman Jerome Powell will kill the proposal. That would probably be too late for Goldman, which heavily cut back on commodities inventory in 2017, and for the reputational reasons outlined above, might never return even if it was allowed to.