[throwit1979]

That's not an answer to the question, though. I'm also genuinely curious. What do people "in finance" actually do on a day to day basis? My naive understanding is that the industry is highly automated already, so what activities are these 100 hour weeks spent doing?

[cmbaus]

I've spent most of my career servicing the financial services and investment industry. While I've enjoyed much of it, I also found a lot of it surprising to say the least.

What many people in finance haven't realized is most of the industry is an IT problem. Much of the work I see involves ingesting data, analysing it, and reporting on it. Most everything else is sales.

Most of the people doing this work come from a finance background and only have a rudimentary understanding of IT or programming. Many know Excel and PowerPoint pretty well, and those tools are pushed to extremes.

The work I see being done by junior analysts is often manual IT work. You wouldn't believe what goes on behind the scenes assembling the statements and reports that are sent to clients. It can be surprisingly difficult to sell software in this environment, because many stakeholders would rather get another junior analyst than solve the problem with software. There are some enlightened firms when it comes to technology, but I would say that is the exception.

Ultimately what I think is going to happen is that companies that are built from the ground up around a technology infrastructure will become more efficient and win business from traditional companies. It will take a long time to displace the bigger players, but it will happen. WealthFront is a good example of a financial company that is built around technology. Eventually these types of firms will win out in the market.

Update: Here is an example of the type of work I see analysts doing:

Manually copy data from 5 different systems into a spreadsheet at the end of the month. The data always has flaws, so go back and validate the data and resolve the issues. When the data is clean, do some processing on the data in the spreadsheet and create some graphs.

Then cut and paste output from the Excel spreadsheet into 150 client PowerPoint presentations, and combine that with commentary saved in an Word document from another department that is stored on a network drive.

At the end of the month an analyst will work 80 hour weeks to complete this in 3-5 business days. This type of workflow is not uncommon.

[maqr]

The cynic in me wonders if the inefficiency and inaccuracy that comes with using humans to crunch numbers gives them more room for error. It's pretty hard to pressure a computer to make the numbers work in your favor, but it's pretty easy to push a low level employee to do something unethical.

[cmbaus]

It might happen, but at least in the segments of the industry I work in, I don't think it is common. What is more common is human error. Most of the businesses I deal with are highly regulated, and they wouldn't risk fudging the numbers. In fact much of our business these days is related to enforcing compliance regulations.

If you knew how dirty most financial data is when it is released, you would probably be checking your statements more carefully.

[eru]

Oh, a certain level of human error amounts to the same thing as fudging. Because only errors in the unwanted direction will be chased.

[mtew38]

Investment bankers are masters of Excel, Powerpoint, and Word and they're messing around in one of those all day long. Mostly building financial models and writing or formatting presentations. The work to do is endless, and if it's not the senior members of a team will make something up for you to do.

A huge part of what i-bankers do is pitch - either pitch to a company that you'd like to bring on as a client, or pitch to potential buyers of your client (and variations of that). Junior guys basically make the presentations for those pitches. There's generally horrible project management going on at the senior level as well, and senior team members have no qualms about making you make pointless updates to a presentation (for example, update the numbers in the presentation for the stock price today even though nobody will see the presentation for a week, and you'll have to update it then anyway). In fact, it's sort of part of the deal.

[lesterbuck]

I guess showing them a live data source for the stock price in the presentation would just make their heads explode. I thought that was a reasonably out-of-the-box Office feature?

[phyalow]

Unlikely, most data points in presentations are specifically given to paint a certain view! Last weeks figure will be last weeks low or high, not some arbitrary figure pulled from the middle of a session...

[cmbaus]

This closely aligns with what I've seen in the field.

[etherael]

The OP answer reads to me like a very polite and indirect way of saying "partially self-inflicted pointless busywork".

[mathattack]

Speaking of what many consider classic investment banking, or Mergers & Acquisitions: - Senior People (Managing Directors and Senior VPs) meet with clients during the day, figure out what to pitch, and what deals need to be done. - Mid-level Managers (VPs) divide up the work ("create this pitchbook", "model this stock price", "come up with values of comparable companies"), do some of it themselves, assign work out, and check it. - Low level folks (Analysts & Associates) spend a lot of time waiting for work, and then execute it until it's done. - Interns try to impress that they have the stamina to get a lot of work done.

There's a lot of waiting implied in this for the juniors that are doing 100 hour weeks. For up to half the day they're goofing off, surfing the net, and looking busy. Then they crunch spreadsheets, and write pitch books.

Other divisions have similar hierarchies, but a little less busywork. Research spends a lot of time writing models and coming up with research papers, pitching trade ideas.

In trading, the hours aren't necessarily that bad, though they can be for juniors. Trading hours are 8-10 hours a day, but the jobs are hard to get so juniors spend a lot of time trying to look and impress.

[sliverstorm]

I can only imagine they are like programmers, except instead of developing code they are developing models on which to base decisions.