What trading jobs at JPMorgan, Goldman Sachs, Morgan Stanley, Citi and BofA are really like

So you want to work in trading for a U.S. investment bank? Choose your spot with care: one bank's approach to risk is very different to another's.

The discrepancy between U.S. banks' attitudes to trading is amply illustrated by the 10-K reports recently filed by JPMorgan, Morgan Stanley, Goldman Sachs, Citi and Bank of America for 2018.  Within these reports are charts reflecting banks' approach to market risk and the gains and losses their traders made in 2018. These charts, we have replicated below. 

They make it very clear that while some banks (JPMorgan) like taking risk and often make losses (JPM fourth quarter), others (Bank of America, Citi) are extremely careful, with the result that their traders make very few losses at all. When you're choosing where to work as a trader, you might want to take this into account: after all, not everyone is suited to an environment of cautious risk aversion or of wild swings.

JPMorgan's risk-taking traders: Big gains, big losses 

JPMorgan's 10-K suggests it's a place for risk fiends. In 2018 the bank's traders had 128 days of gains and 151 days of losses. As the chart below, the final quarter seems to have been a particulary messy one for JPM's traders, with endless loss making days and hardly any gains. 

This doesn't seem to bode well for trading jobs at the bank - particularly given that trading revenues there are down in the "high teens" in percentage terms in the first quarter.  However, there's been no news of layoffs yet.

JPMorgan Daily Market Risk-Related Gains and Losses vs. Risk Management VaR (1-day, 95% Confidence level):

Goldman Sachs' careful risk takers: Many big gains, a few big losses 

Much has been made of Goldman traders' big fourth quarter loss. Alongside JPMorgan's traders, however, they look like elderly citizens on mobility scooters. Goldman's traders made losses on 31 days in 2018. On 12 days, they made gains in excess of $100m - while JPMorgan's traders only achieved this on one day last year. 

Goldman Sachs' frequency distribution of daily net revenues for positions included in VaR for 2018:

Morgan Stanley's moderately cautious traders: Few big losses, few big gains

Morgan Stanley's traders are more cautious again. As the chart below shows, they were only loss-making on 16 days last year. However, they also made fewer big gains than traders at Goldman Sachs (although more than traders at JPMorgan) and say more of their returns clustered between $25m and $75m per day.

Bank of America's very careful traders: No big losses, very few big gains

BofA's traders take 'careful' to a whole new level. Last year, they only made losses on five days (remember JPMorgan?). On no days at all did BofA's traders make losses in excess of $25m. Although they increased the number of days on which they made gains between $75m and $100m compared to 2017, there were only around two days when they made more money than this. 

Citi's equally careful traders: No big losses, but a few giant gains 

Lastly, there's Citi. Like the traders at BofA, Citi's traders only made losses on five days in 2018. However, their fat tails were distinctly fatter than their closest rivals'. - There was one day when Citi's traders made losses of between $30m and $40m. And there were around 14 others when Citi's traders made gains of between $120m and $160m. Citi's traders look like a more moderate version of Goldman Sachs: once in a while, they too make giant gains, but they're less prone to offsetting this with giant swings in the opposite direction. 

Have a confidential story, tip, or comment you’d like to share? Contact: sbutcher@efinancialcareers.com in the first instance. Whatsapp/Signal/Telegram also available. Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)