LBS is currently experiencing an upside breakout in inbound financial modeling questions.
I recently had an old friend from business school over to visit, who happens to be a sell side analyst, and while we were outside enjoying a cigar on an unseasonably warm day he asked me:
“How do buy-siders use sell side models?”
Then, the very next day, a loyal reader of LBS asked if I would write an article exploring the differences between buy side and sell side financial modeling.
OK, I get it. Time to do an article on financial modeling.
“What Gets Measured Gets Managed”
There. I said it.
It’s true. There’s even academic research that supports the case that sell side earnings estimates (the primary output of financial models) are better than buy side analyst estimates.
Score one for the sell side.
Does this mean sell side analysts are also better stockpickers, better at analyzing data, and better looking?
No, no, and … well, I guess it’s possible they are better looking since they have to sell their research to clients (can we start calling them research whores?).
Anyway, the reason the sell side is better at modeling has to with something management guru Peter Drucker stated years ago: “What gets measured gets managed.”
On the sell side, a primary job function is to generate earnings estimates that become part of the “consensus” estimates (Note: The consensus estimate is simply the average of all sell side analyst estimates) you hear so much about during earnings season. Since they are so public, sell side analysts really want their estimates to be right.
Imagining sharing your personal goals with the world on Facebook, the tell me that wouldn't create a lot of pressure to follow-through.
What gets measured gets managed.
On the buy side, a primary job function is not the generation of earnings estimates, but rather the analysis of those estimates. Buy side analysts don't get paid to produce estimates, they get paid to produce alpha.
Since they aren't compensated by producing earnings estimates, buy side analysts tend to rely on sell side models for their analysis.
The Never-Ending Conference Call
As you may know, most companies hold quarterly conference calls to discuss their earnings results with analysts. Sell side analysts typically ask most of the questions, but buy side investors are always listening as well.
In a typical conference call, the management team will usually go over some prepared remarks that are usually just a regurgitation of their earnings press release.
(Note to all management teams everywhere: Both buy side and sell side analysts can read. Accept this fact and give us something new … we read the press release last night.)
After the prepared remarks, the management team will typically open up the call to Q&A where sell side analysts take turns asking questions. At first, these questions usually focus on company specific issues that can drive investment performance such as industry or company trends. But then they inevitably devolve into questions about modeling the company's financials.
“How should I think about modeling depreciation over the next year?”
“What tax rate is embedded in your guidance assumptions?”
“How should we model [insert obscure line item here] when thinking about [insert pointless circumstance here]?”
At this stage in the call, most buy side analysts resort to trolling the Internet for anything to make the pain of boredom stop.
If your entire investment thesis hinges on whether a company's tax rate is 32.1% versus 33.3%, then you have officially lost the forest for the tress.
Pennies don't move stocks.
OK, pennies can move penny stocks, but you get my point.
In order to convince your portfolio manager of an investment recommendation, you need to be convincing. Sell the story, sell the analysis … it needs to be so clear and obvious that your PM absolutely has to follow your recommendation.
Sell side analysts need to get their models right, so you don't need to worry about forecasting the correct tax rate. Just use the sell side models and enter in your own assumptions, if you must.
You're paying the sell side brokerage commissions to get their equity research, you might as well use it.
It Pays to Be Contrarian … Except When it Doesn't
Sell side analysts face a different set of circumstances when dealing with pennies. For the sell side, pennies do matter.
But did you not just say pennies don't matter?
Yes, I did. Before you accuse me of losing it (“Mike has gone bye-bye, Egon”), I need to further explain the role of sell side estimates.
Earnings results are important indicators of a company's health. They are the E in the venerable P/E ratio so important to value investors everywhere.
But good earnings don't mean jack if everyone is expecting good earnings. A good company doesn't always make a good investment. Imagine a scenario where a company has 10 sell side analysts covering it and all 10 analysts project the company to do $1.50 in earnings per share next quarter. If the company reports earnings of $1.50, what do you think will happen to the stock price?
Probably nothing, since everyone expects it to happen.
Now, imagine instead the company reports $1.00 in earnings and cites deteriorating demand as the reason for the miss.
The stock will tank and, if you are long that stock, you are going to have a very bad day.
Being consensus will get a buy side analyst exactly nowhere. There just aren't many alpha opportunities following the consensus.
Buy side analysts are always looking for alpha-generating ideas. One way is to examine sell side earnings estimates and take a closer look at the outlier estimates among the sell side analysts.
If an analyst is way below the consensus estimate, it could mean there's a short opportunity potentially not priced. But only if that estimate proves to be correct …
Sell side analysts get paid for being right. They need to get their models to work. If a PM shorts a stock based on your below consensus estimate but the actual results end up exceeding consensus, that PM is going to be upset and the sell side analyst just lost some credibility.
Get the models right on the sell side and use the right models on the buy side.
All Forecasts are Foolish
Of course, none of this really matters since all forecasts are shit. No matter how awesome your model, you will be wrong. Various analyses have shown that professional forecasts, be that for economic forecasts or S&P 500 earnings, all tend to lag the real results.
In other words, analysts are very good at forecasting exactly what has just happened. Reality never seems to go as planned. Just remember that you will be wrong, just don't look like a fool doing so by being the outlier on a forecast and being wrong at the same time. Make sure your analysis is thorough.
About that Buy Side vs. Sell Side Study …
I mentioned a study earlier that compared buy side analysts to sell side analysts. An interesting conclusion from the study was that sell side stock picks outperformed buy side stock picks.
Before the sell-siders out there rejoice and tell the the buy-siders to “Suck it,” don't forget who pays your bonus!
Seriously, though, despite the authors' best efforts it is nearly impossible to create an apple-to-apples comparison of buy side versus sell side stock picks. That said, the study did find that sell-siders who jump to the buy side are still able to generate alpha.
This is interesting and very important point for aspiring buy side analysts. Going to the sell side first is a very good option for many people.
In fact, I might even put it as my preferred path to the buy side. The benefits are meaningful:
- More Options: There are more sell side analysts/associates than there are buy side analysts. This means there are more opportunities to find an entry-level position in equity research.
- Better Modeling Skills: If you desire to become great at financial modeling, learning on the sell side (either equity research or investment banking) is the way to go.
- Better Compensation: Sell side comp is typically comparable to or better than the buy side. Remember that the big money on the buy side only happens when you start your own hedge fund.
- Better Investment Feedback: A buy side analyst has only his PM to investment ideas off. Sell side analysts are always talking to their dozens or even hundreds of clients.
- Better Networking: Buy side analysts typically are most often in contact with the sell side analysts they follow and other buy side analysts who cover the same companies. Sell side analysts will typically meet with a broader range of buy side analysts and PMs since they have hundreds of clients. This is why the typical exit opportunity for the sell side is to the buy side.
That's a pretty convincing set of benefits.
But You Said to Go Straight to the Buy Side?!?
I've mentioned before that you don't necessarily need to take the almost-cliche investment banking route to the buy side, but that doesn't preclude starting in equity research on the sell side.
Entry-level level jobs in asset management or hedge funds are nearly identical to entry-level equity research jobs.
So, if you want to go to the buy side, by all means search for buy side jobs. But also include equity research jobs at the top of your list also. Unless you have a target school degree, you've completed all three levels of the CFA, and your dad is an MD at Goldman, it's best to keep all of your options open.
Especially in this tough economy for finance jobs.