How Serious Are Analysts Conflicts of Interest Leslie

How Serious Are Analysts’ Conflicts of Interest? Leslie Boni, University of New Mexico Kent L. Womack, Tuck School at Dartmouth 1

So Far: Hearings, Settlements, and Rules Changes Summer 2001 Congressional Hearings: “Analyzing the Analysts” » May 2002 Merrill Lynch $100 Million Settlement with NY State » Summer 2002 Agrees to change how analysts monitored and paid New NASD Rule 2177 & NYSE Rule 472: » » 2003 Ongoing Complaints from retail investors post-market highs Separates Research and Investment Banking Prohibits companies from reviewing own ratings/targets Prohibits offering favorable research for I/B business Increases disclosure (incl. personal trading and history) “Global Research Settlement” with 12(? ) Large Firms » » $1. 4 Billion + in fines and funds for restitution, independent research, & education Elimination of analyst compensation from I/B and participation in road shows 2

What’s the Academic Literature Say about the Conflicts? • Affiliated investment bankers are overly optimistic at IPO – Michaely and Womack (1999): “Conflict of Interest and the Credibility of Underwriter Analyst Recommendations, ” Review of Financial Studies • Analysts are an important factor when companies pick bankers – Krigman, Shaw, and Womack (2001): “Why Do Firms Switch Underwriters? ” Journal of Financial Economics • Analysts are optimistic to obtain access to management – Lim (2001): “Rationality and Analysts’ Forecast Bias, ” Journal of Finance • Optimistic analysts are more likely to move to better firms – Hong and Kubik (2003): “Analyzing the Analysts: Career Concerns and Biased Earnings Forecasts. ” Journal of Finance 3

“Analysts, Industries, and Price Momentum” (Boni and Womack, 2003) Can analysts rank future winners and losers in their industry? Ø Ø Analysts are industry specialists. Proper test of stock-picking ability is an industry-based analysis. Analysts signal rankings with upgrades and downgrades. Examine self-financing portfolios (long net upgraded stocks, short net downgraded stocks). Dataset: Ø U. S. stocks. ~150, 000 recommendations on 7, 766 companies from 433 brokerages. Time period: Jan. 1996 – June 2001. Findings: Ø Ø Analysts can rank stocks: Returns 1. 4% per mo. , 18% per yr. Returns are better if stocks have fewer analysts covering them. 4

“Analysts, Industries, and Price Momentum” (Boni and Womack, 2003) Recommendation-Change Portfolios Mean = 1. 4% Standard deviation = Sharpe Ratio = 0. 57 1. 7% Jegadeesh and Titman (1993) Price Momentum Portfolios Mean = 1. 3% Standard deviation = 11. 6% Sharpe Ratio = 0. 08 5

The Value of Analyst Research: Retail versus Institutional Investors’ Perspectives Boni and Womack (2003): Financial Analysts Journal, forthcoming How do we reconcile: Ø Institutional investors say they still value analysts’ research despite the potential for conflicts of interest. Ø Retail investors complained analysts didn’t get them out of tech stocks post-market highs. Dataset: Ø U. S. stocks. ~150, 000 recommendations on 7, 766 companies from 433 brokerages. Ø Time period: Jan. 1996 – June 2001 Ø About 50% of recommendations are from largest 25 firms Ø Industry categories: S&P/Morgan Stanley GICS codes. 6

Transition Matrix of Analysts’ Recommendations: I/B/E/S Data for 1996 through 1 st Half 2001 Downgrades ~ 50% Upgrades ~ 50% “Buy” = 67% “Hold” = 30% “Sell” = 3% 7

Figure 1 Consensus Levels: Software Industry and Microsoft 8

Returns from U. S. Stocks: Retail Investor’s Perspective (1996 through 1 st Half 2001) 9

Returns from U. S. Stocks: Inst’l Investor’s Perspective (1996 through 1 st Half 2001) 10

Value of Analysts’ Recommendations (1996 – 2 Q 2001) Analysts on average are industry experts. • Buying all stocks with analyst “buy” consensus levels beat the S&P Index. • Buying upgraded stocks was better! • Buying upgraded stocks & short selling downgraded stocks was much better! “Tech-only” strategies: • Buying consensus levels or upgrades beat the S&P on average … but lost big from 2 Q 2000 on. • Buying tech upgrades & short selling tech downgrades was a big winner even from 2 Q 2000 on! Ø Tech stocks harder to analyze? Conflicts more severe? Or Both? 11

Did Analysts’ Recommendations Encourage Investors to Overweight Tech Stocks? 12

Is independent research better? Is brokerage without investment banking better? Brokerage with I/B Brokerage Independent Only Research Pressure for “Optimism” from: • Investment Banking - - • Companies Covered ? • Buy-Side Clients ? • Brokerage Commissions ? • Firm’s Trading Positions ? • Analyst’s Trading Positions ? 13

New rules and settlements go too far? Not far enough? • “Fixes” to date don’t eliminate many of the pressures for “optimism”. Ø Unclear why independent research or brokerage w/o I/B will be better. • Research budgets cut. Analysts cut. Reduced coverage or coverage eliminated for some Industries and stocks. Ø Some companies will have to purchase coverage. Ø Need greater revenues from brokerage commissions. Churning? Ø Reduced competition by analysts increases value of recommendations. • Problems if brokerage firms choose independent research to fund: Ø Why fund independents with lower buy: sell ratio than own firm? Ø Why fund independents that are better at stock picking than own firm? Ø Good independents will realize this and not go after funding/linkages? Ø Retail investors will continue to be at a disadvantage!? !? !? 14