Figure 1: Distribution most recent forecasts of annual earnings pe r share [1980-2006] N=790,718, among which 49.55% end with a “0” or “5”.
9, 000 8, 000 7, 000 6, 000
frequency
5, 000 4, 000 3, 000 2, 000 1, 000 0 - 2. 5 -2 - 1. 5 -1 - 0. 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 6 Recent FEPS1
[1980-1994] N=324,291, among which 67.69% end with a “0” or “5”.
6, 000 5, 000
frequency
4, 000 3, 000 2, 000 1, 000 0 - 2. 5 - 2 - 1. 5 - 1 - 0. 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 6 Recent FEPS1
[1995-2006] N=466,427, among which 36.93% end with a “0” 0r “5”.
4, 000 3, 500 3, 000
frequency
2, 500 2, 000 1, 500 1, 000 500 0 - 2. 5 -2 - 1. 5 -1 - 0. 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 6 Recent FEPS1
1
Figure 2: Analysis over time of the most recent forecast of annual EPS (FEPS1) for each analysts (the forecast made closest to the earnings announcement) A. Percentage of zero/five FEPS1 forecasts over time
0. 7
Percentage of zero/five recent forecasts
0. 6 0. 5 0. 4 0. 3 0. 2 0. 1
B. Accuracy analysis: Average absolute value of forecast errors (FE_P) for zero/five and other forecasts, and the difference (FEPS1 sample)
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
zero/five other Difference
0
-0.01
C. Bias (optimism): Average forecast error (FE_P) for zero/five and other forecasts, and the difference (FEPS1 sample)
0. 06 0. 05 0. 04 0. 03 0. 02 0. 01
- 0. 01
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
zer o/ f i ve ot her di f f er ence
0
2006
2
Figure 3: Distribution all two-year ahead forecasts of annual earnings per share (FEPS2) [1980-2006] N=1,862,597, among which 60.88% end with a “0” or “5”
30,000 25,000 20,000
frequency
15,000 10,000 5,000 0 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 FEPS2
[1980-1994] N=600,230, among which 76.11% end with a “0” or “5”
14, 000 12, 000 10, 000
frequency
8, 000 6, 000 4, 000 2, 000 0 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 FEPS2
[1995-2006] N=1,262,367, among which 53.64% end with a “0” or “5”.
16, 000 14, 000 12, 000 10, 000 8, 000 6, 000 4, 000 2, 000 0 - 2. 5 -2 - 1. 5 - 1 - 0. 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 6
frequency
FEPS2
3
Figure 4: Analysis over time of two-year ahead forecasts of earnings per share (FEPS2) A. Percentage of zero/five FEPS2 forecasts over time
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Percentage of zero/five FEPS2
1984
1986
1990
1992
1994
1996
1998
2000
2002
2004
2004 2005
B. Accuracy analysis: Average absolute value of forecast errors (FE_P) for zero/five and other forecasts, and the difference (FEPS2 sample)
0.2 0.15 0.1 0.05 0
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
-0.05
zero/five
other
Difference
C. Bias (optimism): Average forecast error (FE_P) for zero/five and other forecasts, and t he difference (FEPS2 sample)
0. 2
0. 15
0. 1
0. 05
1984
1986
1988
1989
1991
1993
1995
1996
1998
2000
2003
2005
1985
1987
1990
1992
1994
1997
1999
2001
2002
2004
- 0. 05
z er o/ f i ve
ot her
Di f f er ence
2006
0
2006
2006
1988
4
Figure 5: Distribution long-te rm earnings growth forecasts [1983-2006] N=381,960, among which 44.2% end with “0” or “5”
45,000 40,000 35,000 30,000
frequency
25,000 20,000 15,000 10,000 5,000 0
0
5
10
15
20
25
30
35
40
45
LTG forecast
[1983-1994] N=183,148, among which 37.9% end with “0” or “5”
20,000 18,000 16,000 14,000
frequency
12,000 10,000 8,000 6,000 4,000 2,000 0
10
15
20
25
30
35
40
45
LTG forecast
[1995-2006] N=198,812, among which about 51.1% end with “0” or “5”
25,000 20,000
frequency
15,000 10,000 5,000 0
10
15
20
25
30
35
40
45
LTG forecast
50
0
-5
5
50
0
-5
5
50
-5
5
Figure 6: Analysis over time of long term growth forecasts (LTG) A. Percentage of zero/five LTG growth forecasts over time
0. 6 0. 5 0. 4 0. 3 0. 2 0. 1 0
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
B. Accuracy analysis: Average absolute value of forecast errors for zero/five and other forecasts, and the difference (LTG forecast sample)
30 25 20 15 10 5 0
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Zero/five
Other
Difference
C. Bias (optimism): Average forecast error for zero/five and other forecasts, and the difference (LTG sample)
20 18 16 14 12 10 8 6 4 2 0 -2
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
zero/five
other
Difference
2002
2002
2006
6
Figure 7: Distribution management guidance (point forecasts of annual EPS) [1994-2006] N=4,270, among which 46.79% end with a “0” or “5”
100 90 80 70
frequency
60 50 40 30 20 10 0 - 2. 5 -2 - 1. 5 -1 - 0. 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 6
[1994-2001] N=1,666, among which 48.55% end with “0” or “5”
35 30 25
frequency
20 15 10 5 0 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
[2002-2006] N=2,604, among which about 45.66% end with “0” or “5”
70 60 50
frequency
40 30 20 10 0 - 2. 5
-2
- 1. 5
-1
- 0. 5
0
0. 5
1
1. 5
2
2. 5
3
3. 5
4
4. 5
5
5. 5
6
7
Figure 8: Analysis over time of management guidance (point forecasts of annual EPS) A. Percentage of zero/five management forecasts over time
0.6 0.5 0.4 0.3 0.2 0.1 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
B. Accuracy analysis: Average absolute value of forecast errors for zero/five and other forecasts, and the difference (Management guidance sample)
0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 -0.01 -0.02 zero/five other difference 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
C. Bias (optimism): Average forecast error for zero/five and other forecasts, and the difference (Management guidance sample)
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 -0.01 -0.02 zero/five other difference 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
8
Figure 9A: Average size-adjusted returns to three portfolios based on the percentage of 0 and 5 long-term growth forecasts
3%
2%
2%
1%
1%
0% PCT_05 < 25% -1% 25%
75%
-1%
Figure 9B: Twelve-month hedge returns to LTG PCT_05 strategy over time
30% 20% 10% 0%
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
12
- 10% - 20% - 30%
Figure 9C: Cumulative analyst LTG forecast revisions over the twelve months after the first fiscal quarter end for the three PCT_05 groups
Cumulative LTG forecast revisions
0 -0.2 -0.4 -0.6 -0.8 -1 -1.2 -1.4 -1.6 -1.8 -2
1
2
3
4
5
6
7
8
9
10
11
Month relative to portfolio formation date
PCT_05 =<25% 25%75%
Note: Hedge returns equal to the difference of average twelve-month size -adjusted returns between the low and high PCT_05 groups. PCT_05 <25% has 10,593 observations, 25% 75% has 15,986 observations.
2004
9
Exhibit 1A: Timeline of short-term return tests
Use the most recent EPS forecasts made within 10 to 365 days before earnings announcement to calculate percentage of 0 and 5 forecasts Three days returns around earnings announcements
-365
-10
-1
0
1
Earnings Announcement
Exhibit 1B: Timeline of long-term growth in EPS tests for a December 2000 firm-year observation
Fiscal year end 12/31/00 (Twelve-month return period) Q1 of year 02 Fiscal year end 12/31/05
Q1
Q1
(the most recent LTG forecasts made by each analyst within 365 days)
(Five years)
Consensus and the median of all recent LTG forecast: LTGCF
Actual past five-year growth by IBES: LTG
Forecast errors = LTGCF-LTG
10