Some Facts About the Tech Bubble in the Late 1990s
(Some of the Slides from a Recent NBER Discussion )
Pietro Veronesi Graduate School of Business, University of Chicago CEPR, NBER
NBER – April 2006
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
Some Facts About the Nasdaq “Bubble” A. What “went up” at the end of the 1990s?
page: 1
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 1
Some Facts About the Nasdaq “Bubble” A. What “went up” at the end of the 1990s? 1. Nasdaq Index.
Cumulative Return: NYSE and NASDAQ 4
3.5
3
2.5 NYSE 2
1.5
1
0.5
0 1997
1998
1999
2000
2001
2002
2003
2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 2
Some Facts About the Nasdaq “Bubble” A. What “went up” at the end of the 1990s? 1. Nasdaq Index. 2. Nasdaq Volatility. Nasdaq Volatility 4
0.5
3.5
0.45 Volatility
3
0.4
2.5
0.35
2
0.3
1.5
0.25
1
0.2
0.5
1997
0.15
1998
1999
2000
2001
2002
2003
2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 3
Some Facts About the Nasdaq “Bubble” A. What “went up” at the end of the 1990s? 1. Nasdaq Index. 2. Nasdaq Volatility. 3. Nasdaq Beta. Nasdaq Beta 4
2
Beta 1.75
1.5
2
1.25
1
0 1997
1998
1999
2000
2001
2002
2003
0.75 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
The Less Famous NYSE High Tech “Bubble” B. Whatever happened to Nasdaq also happened to NYSE High-Tech Stocks.
page: 4
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 4
The Less Famous NYSE High Tech “Bubble” B. Whatever happened to Nasdaq also happened to NYSE High-Tech Stocks. – NYSE Hi Tech Stocks went up compared to NYSE Lo Tech Stocks
Cumulative Return: NYSE High Tech, NYSE Low Tech and NASDAQ 4 NYSE High Tech NYSE Low Tech Nasdaq 3.5
3
2.5
2
1.5
1
0.5 1997
1998
1999
2000
2001
2002
2003
2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 5
The Less Famous NYSE High Tech “Bubble” B. Whatever happened to Nasdaq also happened to NYSE High-Tech Stocks. – NYSE Hi Tech Stocks went up compared to NYSE Lo Tech Stocks – NYSE Hi-Tech Volatility has the same properties of Nasdaq Volatility NYSE Hi−Tech Volatility 3
0.35
0.3
2.5
2
0.25 Volatility
1.5
0.2
1
0.5 1997
0.15
1998
1999
2000
2001
2002
2003
0.1 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 6
The Less Famous NYSE High Tech “Bubble” B. Whatever happened to Nasdaq also happened to NYSE High-Tech Stocks. – NYSE Hi Tech Stocks went up compared to NYSE Lo Tech Stocks – NYSE Hi-Tech Volatility has the same properties of Nasdaq Volatility – NYSE Hi-Tech Beta has the same properties of Nasdaq Beta. NYSE Hi−Tech Beta 3
3
Beta 2.5
2.5
2
2
1.5
1.5
1
0.5 1997
1
1998
1999
2000
2001
2002
2003
0.5 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 7
Trading Volume in Tech Stocks C. Is it really true that trading volume in tech stocks went up in the late 1990s?
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 7
Trading Volume in Tech Stocks C. Is it really true that trading volume in tech stocks went up in the late 1990s? – Nasdaq Turnover increased somewhat. ∗ But it peaked well after the “bubble” burst.
Nasdaq Turnover 4
0.5
Turnover
3.5
3
0.4
2.5
0.35
2
0.3
1.5
0.25
1
0.5 1997
0.45
0.2
1998
1999
2000
2001
2002
2003
0.15 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 8
Trading Volume in Tech Stocks C. Is it really true that trading volume in tech stocks went up in the late 1990s? – Nasdaq Turnover increased somewhat. ∗ But it peaked well after the “bubble” burst. – NYSE Hi Tech Stocks’ Turnover did not move at all NYSE Hi Tech Turnover 3
0.18
Turnover 2.5
0.16
2
0.14
1.5
0.12
1
0.5 1997
0.1
1998
1999
2000
2001
2002
2003
0.08 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 9
Trading Volume in Tech Stocks C. Is it really true that trading volume in tech stocks went up in the late 1990s? – Nasdaq Turnover increased somewhat. ∗ But it peaked well after the “bubble” burst. – NYSE Hi Tech Stocks’ Turnover did not move at all – Nasdaq Hi Tech Stocks’ Turnover went down. Nasdaq Hi Tech Turnover 4.5
0.5
4
0.45
3.5
Turnover
3
0.35
2.5
0.3
2
0.25
1.5
0.2
1
0.5 1997
0.4
0.15
1998
1999
2000
2001
2002
2003
0.1 2004
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
Why Did the “Bubble” Burst? D. There is little doubt of what caused tech stock prices to drop in 2000.
page: 10
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 10
Why Did the “Bubble” Burst? D. There is little doubt of what caused tech stock prices to drop in 2000. • Nasdaq’s Profitability plummeted in 2000. Nasdaq and NYSE/Amex M/B ratios 7
NYSE/Amex Nasdaq
6
5
M/B
4
3
2
1
0 1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
Nasdaq and NYSE/Amex profitability 20
NYSE/Amex Nasdaq
15
Return on equity, percent per year
10
5
0
−5
−10
−15
−20 1960
1965
1970
1975
1980
1985
1990
1995
(Source: Pastor and Veronesi “Was There a Nasdaq Bubble in the Late 1990s?” JFE, 2006)
2000
2005
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 11
Was it a Bubble? • It is possible that prices exceeded fundamentals in the late 1990s for tech stocks. • However, an explanation based on short sales constraints may find it hard to explain why exactly the same pattern of return, volatility and “risk” occurred in NYSE Hi Tech stocks. – Short sales constraints are much less binding for NYSE stocks. • Short sales constraints though may have played a role in Nasdaq – The “bubble” was stronger in Nasdaq – Trading seemed more active in Nasdaq during the “bubble” period (but it peaked later).
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 12
Was it a Bubble? • Uncertainty about long term growth and low risk aversion may also have played an important role. (see Pastor and Veronesi “Was there a Nasdaq bubble in the late 1990s?”, 2006, JFE)
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 12
Was it a Bubble? • Uncertainty about long term growth and low risk aversion may also have played an important role. (see Pastor and Veronesi “Was there a Nasdaq bubble in the late 1990s?”, 2006, JFE)
– This story explains all of the facts discussed earlier. 1. High uncertainty about long term growth + low risk aversion imply (very) high prices. 2. High uncertainty implies a high return volatility and a high beta. 3. High uncertainty implies a strong reaction of prices to bad news in profitability (and hence price drop) 4. High uncertainty + differences of opinion implies a high trading volume.
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 12
Was it a Bubble? • Uncertainty about long term growth and low risk aversion may also have played an important role. (see Pastor and Veronesi “Was there a Nasdaq bubble in the late 1990s?”, 2006, JFE)
– This story explains all of the facts discussed earlier. 1. High uncertainty about long term growth + low risk aversion imply (very) high prices. 2. High uncertainty implies a high return volatility and a high beta. 3. High uncertainty implies a strong reaction of prices to bad news in profitability (and hence price drop) 4. High uncertainty + differences of opinion implies a high trading volume. – Moreover, it explains other facts about the late 1990s 1. High uncertainty increases option value of new ventures and thus increase incentive to go public =⇒ IPO wave of tech stocks at the end of the 1990s
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution.
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
• Why?
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
• Why? – Productivity of new technologies is uncertain ∗ =⇒ High volatility
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
• Why? – Productivity of new technologies is uncertain ∗ =⇒ High volatility – Initially, this uncertainty is mainly idiosyncratic as new technologies are developed on a small case ∗ =⇒ prices up (e.g. PV 2003)
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
• Why? – Productivity of new technologies is uncertain ∗ =⇒ High volatility – Initially, this uncertainty is mainly idiosyncratic as new technologies are developed on a small case ∗ =⇒ prices up (e.g. PV 2003) – If technology was eventually adopted (as in a Tech Revolution), the uncertainty gradually changes from idiosyncratic to systematic. ∗ =⇒ discount rate up and prices down.
Pietro Veronesi
Some facts about the 1990’s Tech “Bubble”
page: 13
Was it a Bubble? • Uncertainty and learning also imply one additional reason why prices should decline during a technological revolution. – Technological Revolutions imply an Increase in Risk (See Pastor and Veronesi, “Technological Revolutions and Stock Prices,” NBER WP 2005)
• Why? – Productivity of new technologies is uncertain ∗ =⇒ High volatility – Initially, this uncertainty is mainly idiosyncratic as new technologies are developed on a small case ∗ =⇒ prices up (e.g. PV 2003) – If technology was eventually adopted (as in a Tech Revolution), the uncertainty gradually changes from idiosyncratic to systematic. ∗ =⇒ discount rate up and prices down. – This model implies all of the stylized facts discussed earlier.
