How accurate are 3rd party stock analyst reports?

September 27, 2009 10:20 AM

How accurate are third-party stock analyst reports, such as S&P and Argus?

I recently got an account at Schwab, and they provide the following 3rd-party analyst reports:
- Schwab Equity Rating
- Credit Suisse Research
- Ned Davis Research
- Argus 12 Month Rating
- Standard & Poor Stock Reports
- Reuters Research
- Market Edge Second Opinion

I especially like the S&P and Argus reports, because they provide the most background information supporting their positions. My question is, how can I find data on how accurate these reports are? For example, is there a website that tracks 3rd-party predictions vs. results? In lieu of that, is there a way to access older reports from these analysts so I can compile my own data?

Thanks!

(P.S. please no anecdotal info!)

posted by blahtsk to work & money (8 comments total)

Yahoo! has their Analyst Performance Center which lets you search a given stock symbol and view the best-rated stock analysts for that given company. (Just Googled "stock analyst performance", for what it's worth, but remember that analysts are still just taking shots in the dark, too. Some just have a slightly better candle and a bit better night vision than others. "Past performance is not indicative of future results" rings especially true here.)
posted by disillusioned at 12:11 PM on September 27

None of them are worthwhile. Many studies have shown that professional stock pickers who spend all day reading stock analyst reports do no better than the market index on average.
posted by JackFlash at 4:01 PM on September 27

JackFlash is unfortunately correct.
posted by dfriedman at 4:02 PM on September 27

@JackFlash / @dfriedman : Great! Can you source your data?
posted by blahtsk at 5:59 PM on September 27

Stock analysts and stock report publishers gather and present information for your perusal. What has happened to the company, what is happening, whatever data they have available. Read it for the info, digest it, but be skeptical of their recommendations.

Stock analysis and reporting cannot predict the future. If even one of them could, it would change the stock market radically.

Prediction is very difficult, especially about the future. — Niels Bohr
posted by exphysicist345 at 9:09 PM on September 27

You should check out Morningstar to supplement the ones listed above. They give pretty through info from a supposedly independent perspective...
posted by the foreground at 6:53 AM on September 28

1. There are lots of ratings systems that track the performance of a given analyst's recommendations. On a Bloomberg terminal, for example, each analyst that covers a stock is ranked based on the historical total returns (price change + dividends) of their recommendations on that stock, or on their entire coverage universe. There's also a database called StarMine that tracks the accuracy of earnings estimates. One free alternative is the Yahoo Analyst Performance Center which seems to incorporate SM data along with some other factors.

2. I generally agree with JackFlash and dfriedman that buy/sell/hold ratings are worth little to nothing, and with exphys that better investors tend to use reports as a source of information, not for the ratings -- but I admit this is subjective/anecdotal and I'm not sure what would count as proof. There are lots of academic papers on the subject and there may be some kind of meta-analysis that demonstrates this to your satisfaction (I'll post a link if I find one) but from what I've seen they often address related questions, not the one you're asking -- for example, JackFlash I think is referring to studies on mutual fund underperformance, but there are many factors that go into that, so the fact that mutual fund managers underperform the broader market doesn't implicate analyst recommendations in particular.
posted by pete_22 at 10:41 AM on September 28

3. As for the firms on your list, research is a side business for many of them and is run as more of a volume business, i.e. lots of quantitative rankings (of both fundamental and technical valuation metrics) and multifactor "scoring systems" that allow each analyst to cover a lot of stocks. AFAIK Credit Suisse is the only "sell side" firm on there (i.e. they make money underwriting/trading the stocks) which is a greater conflict of interest but even post-Blodget it's also a higher-profile, more lucrative job that attracts slightly better candidates (CS is much more likely to hire an equity analyst away from Argus than vice versa) and they have a smaller coverage universe with a tighter industry focus.

4. IMO the best research firms (prestigious and few/no conflicts) are boutiques that focus on a particular industry and get paid through trading commissions or on a subscription basis by exclusively institutional clients, and occasionally for consulting/advisory work. An example would be the new firm that Meredith Whitney is starting to focus on banks/financials. There are a few of these firms with a convincing (to me) ratings track record, but unfortunately, this kind of product is never available to retail investors -- and honestly even with these firms the clients are paying for more in-depth analysis as much as or more than ratings.
posted by pete_22 at 11:02 AM on September 28