Printer Friendly

Deregulation and risk.

The recent market turmoil has brought attention to how deregulation of the financial sector may affect risk. The purpose of our study is to examine the market's perception of risk associated with deregulation. We accomplish this by decomposing security risk around deregulation into systematic and unsystematic risk. We examine deregulation of several industries and find a consistent pattern of risk adjustment to deregulation, whereby the increase in security risk is temporary and largely unsystematic.

**********

The recent events in the financial markets focus attention on the issue of deregulation and the risks associated with deregulation. The intent of deregulation is generally to encourage competition. In theory, deregulation should open an industry to competition, and this competition should stimulate innovation and the development of products that benefit consumers. The increase in competition should also affect profitability, as deregulation lowers barriers to entry and more firms enter the industry.

Based on economic theory, the effect of deregulation on the risk of affected firms is not clear. One argument is that the existence of regulation provides a buffer against competitive pressures from supply and demand, and hence, when regulated firms are no longer regulated, the variability of profits and stock returns should increase. This argument is the buffering theory, based on the idea that regulation protects firms from the effects of competition. (1) For example, prior to the deregulation of the airline industry, airlines, with oversight by the Civil Aeronautics Board (CAB), would set fares and divide routes based on negotiation with one another. (2) After deregulation, the airlines faced competing with each other and with new, start-up airlines.

Another argument is that regulation increases a company's risk because the approval process in regulation makes it difficult for a company to respond quickly to changes in the market. (3) This is the regulatory lag explanation. A delay in responding to demand and supply increases the risk that other firms may be able to capitalize on this delay. For example, in the case of AT&T, a regulated monopoly for many years, the process of changing rates was long and arduous. Though this benefitted the regulated company when costs were going down and, hence, there was a lag to lower rates, it harmed it when costs were going up and an increase in rates took a great deal of time. (4)

Still another argument is that the regulated environment that includes rate approval, such as the case of telecommunication firms, tends to discourage investment in higher-risk opportunities This is because customers tend to capture excess profits instead of the shareholders of the regulated company, yet shareholders of the regulated company may bear losses on bad investments that cannot be passed on to customers through higher rates. In other words, shareholders of rate-regulated firms bear the downside risk, but customers and shareholders share the upside potential of investments. Therefore, once deregulated, firms may shift to higher-risk opportunities, with both customers and shareholders sharing both upside and downside risk. (5)

The purpose of our study is to examine the market's perception of risk associated with deregulation. We accomplish this by separating the risk associated with affected companies' stocks' returns around deregulation into idiosyncratic and systematic risk components. We examine deregulation of several industries--airlines, trucking, telecommunication, and financial services--to assure that the phenomena of changing risk is not associated with a single industry's deregulation. Despite these industries being quite different in terms of business risk, we find a pattern of risk adjustment to deregulation that is consistent among these different industries. Our results have implications for managing companies facing deregulation, as well as regulators and government agencies concerned about the effects of deregulation on risk.

One of the more interesting findings is that the increased risk of companies' stock returns is temporary and largely idiosyncratic. In fact, we find that systematic risk tends to decline after deregulation and that total risk declines after three years. In other words, increasing competition does not increase total or systematic risk. Rather, deregulated industries experience temporary increases in idiosyncratic risk and, in general, temporarily reduced systematic risk. The bottom line is that deregulation appears to increase risk, yet this is temporary, with risk returning to pre-deregulation levels within three years of deregulation. One explanation is that it takes companies approximately three years to learn to compete in the deregulated market and for the investors to adjust their expectations regarding these deregulated companies. A contribution of the present study is to offer evidence that reconciles what appears to be mixed evidence from previous evidence regarding deregulation and risk.

We review the extant research on deregulation and risk in Section I and examine the research that examines whether there are changes in idiosyncratic risk associated with specific events or over time in Section II. We present our hypotheses in Section III, and in Section IV we describe the data and methodology that we use to examine whether there is a relation between risk and deregulation. We provide the results of our analysis in Section V and conclude our paper in Section VI.

I. Previous Evidence on Deregulation and Risk

There is little agreement with respect to whether deregulation results in increases or decreases risk among the members of the deregulated industry. Therefore, the effect of deregulation on risk is an empirical issue. Researchers have examined the changes in risk that occurs in the deregulation of the airline, trucking, telecommunication, and financial services, among other industries.

A. The Airline Industry

From 1938 through 1978, the CAB regulated commercial airlines in the United States. In the regulated environment, the CAB regulated fares, routes, and airline schedules. During this period of regulation, not a single new airline entered the industry. (6) Airlines competed primarily by nonprice means, such as providing luxuries for passengers. The United States deregulated the air cargo part of the business in 1977, and the remainder in 1978 with the passage of the Airline Deregulation Act. (7) There were several phases of deregulation, but the most important was the 1978 Act. (8)

There was concern from the airlines that they would have difficulty competing. Once airlines were able to compete, however, the large airlines eliminated routes to smaller cities and concentrated on developing the hub systems, whereas the start-up airlines competed based on price and filled in the gaps in the routes left from the larger airlines. Though deregulation resulted in lower fares and increased revenues for the airlines, the overexpansion and smaller profit margins made the airlines vulnerable to downturns in the economy, with several large airlines experiencing bankruptcy. (9)

Cavarra, Stover, and Allen (1981) argue that systematic risk should increase with airline deregulation. Whereas Banker, Das, and Ou (1997) find that systematic risk increases with airline deregulation, Allen, Cunningham, and Wood (1990) observe that the systematic risk of airlines was lower after deregulation This conflicting evidence is consistent with Cunningham et al. (1988) and Allen, Cunningham, and Wood (1990), who observe that systematic risk increases sharply immediately following the deregulation, but it eventually falls. The results of the empirical estimates of risk suggest that risk changes with the airline industry deregulation, and that systematic risk is lower following deregulation, but not necessarily immediately.

B. The Trucking Industry

Congress passed the Motor Carrier Regulatory Reform and Modernization Act (MCA) in 1980. This act eliminated price controls and barriers to entry in the trucking industry. (10) Another significant aspect of the law was that it introduced competition by price; before the MCA, truck companies passed along all cost increases to customers. Following the signing of the MCA, the Financial Accounting Standards Board issued a new accounting standard for motor carriers that required writing off the unamortized cost of the intangible interstate operating rights that motor carriers had on their books prior to deregulation. (11)

Whereas there is some research on the profitability of the truck industry postderegulation, there is scant research regarding the effects of this deregulation on the risk of the companies in the industry. Morash, Bruning, and McQuin (1981) find that systematic risk decreases for larger carriers but increases for smaller trucking firms, whereas Van Auken and Crum (1985) document no significant change in risk among motor carriers.

C. The Telecommunication Industry

The Communications Act of 1934 establishes the regulation of telecommunication in the United States. (12) Telecommunications was later deregulated in two phases, beginning in the 1980s. The first phase was the break-up of AT&T in 1982, in response to an antitrust action by the US Department of Justice. (13) The result of the case and subsequent settlement was the break-up of AT&T into seven regional holding companies that handled local telephone, with AT&T left with the long-distance line of business.

The second phase began in 1991 with the Senate passage of S. 173, the Telecommunications Equipment Research and Manufacturing Competition Act of 1991 (TERMCA), but more importantly with the Telecommunications Act of 1996. (14) The former act focuses on the actual equipment available to both consumers and businesses. The latter act is more significant because it contains provisions that open up competition through different means, such as cable broadcasting. (15) The impetus for this phase of deregulation was the realization that convergence of the computer, broadcasting, and telephony industries was compromising the effective distribution of telecommunication services. Though companies in this industry are still subject to rate regulation, these companies are able to compete better with broadcasting and cable broadcasting. A potential effect of the second phase of deregulation is that it may stimulate telecommunication firms to diversify lines of business and, hence, reduce total risk.

There is scant research on the effects of deregulation in this industry on risk. Guthrie (2006) argues that the deregulation of the telecommunication industry shifted risk from customers to shareholders, and Chen and Sanger (1985) document that the total risk of AT&T increased, as compared to the prebreakup period, yet most of this was idiosyncratic risk.

D. The Financial Services Industry

Banking deregulation was not through a single law but rather has occurred in phases. The primary deregulation acts within the banking industry are the Depository Institutions Deregulation and Monetary Control Act of 1980 (DIDMCA), the Garn-St. Germain Depository Institutions Act of 1982, and the 1994 Reigle-Neal Interstate Banking and Branching Efficiency Act (RNIBBEA). The DIDMCA phased out interest rate ceilings, whereas the Garn St. Germain Act allowed banks and thrifts to offer money-market accounts to remain competitive with nonbank alternative investments. The Garn St. Germain Act removed restrictions in two areas: lending in real estate and loans to one borrower. With regard to competition across state lines, individual states removed restrictions imposed by the Douglas Amendment to the 1956 Bank Holding Company Act, which prohibited cross-state ownership of banking institutions, until the early 1980s. The RNIBBEA permitted interstate banking, which provided banks an opportunity to diversify their loan portfolios across state lines.

The Financial Services Modernization Act (FSM), more commonly referred to as the Gramm-Leach-Bliley Act, further deregulated the financial services industry. (16) The FSM Act lowers many of the barriers that existed between different financial service businesses erected by the Banking Act of 1933 and the Bank Holding Company Act of 1956, and opens the door for mergers across industries within the financial sector. (17)

Aharony, Saunders, and Swary (1988) observe that total risk increased following the DIDMCA banking deregulation but that systematic risk declined following deregulation. Examining the same deregulation event, Allen and Wilhelm (1988) do not find any change in systematic risk for banking firms, whereas Fraser and Kannan (1990) observe an increase in risk prior to deregulation but no decrease upon deregulation. Saunders and Smirlock (1987) do not find any effect on systematic risk for either banks or securities firms when banks enter the discount brokerage business.

The evidence regarding risk changes associated with the FSM Act is mixed. For example, Strahan and Sufi (2001) document an increase in systematic risk for insurance companies associated with the passage of the FSM Act, whereas Yildirim, Kwag, and Collins (2006) document a decline in risk associated with the FSM Act for commercial banks, investment banks, and insurance companies. Akhigbe and Whyte (2001), focusing on the late 1999 period, document a decrease in market risk for banks, insurance companies, and securities firms, yet an increase in unsystematic risk for banks and insurance companies. (18) The findings of Mamun et al. (2005) and Neale and Peterson (2005) support a reduction in systematic risk accompanying the FSM Act for insurance companies. Specific to idiosyncratic risk, Correa and Suarez (2009) document lower idiosyncratic risk for banking firms following deregulation.

E. Other Industries

Another deregulation situation occurred in the railroad industry, which experienced two waves of deregulation following the bankruptcy of the Penn Central Railroad in 1970. First, the Railroad Revitalization and Regulatory Reform Act of 1976 reduced some of the regulation of the industry, especially with respect to mergers and rates. (19) The Staggers Rail Act of 1980 followed, which deregulated the industry with respect to pricing, timetables, and contracting. (20) It was feared that deregulation may lead to higher rates, though this was not observed to be the case. (21) Peltzman (1976) examines the deregulation of railroads and does not find any change in total or systematic risk following deregulation.

In other industries, the results regarding the effects on risk are mixed. For example, Chen and Sanger (1985) find that the systematic risk of natural gas producers and distributors increased following deregulation. On the other hand, Peltzman (1976) observes that the total and systematic risk of drug companies declined following deregulation.

II. Evidence on Idiosyncratic Risk

Researchers document changes in idiosyncratic volatility, but there is no agreement regarding the source of these changes. For example, Campbell et al. (2001) document an increase in the idiosyncratic volatility of individual company securities but not for the market as a whole. Later research by Brandt et al. (2010), as well as Bekaert, Hodrick, and Zhang (2008), does not support the idea that there is an upward trend in idiosyncratic volatility; rather, individual firm increase in volatility may be episodic in nature.

Explanations for the changes in idiosyncratic volatility suggested by extant research suggest that the increase in idiosyncratic volatility may be due to:

* increasing institutional ownership (Bennett, Sias, and Starks, 2003);

* issuance of equity (Fink et al., 2005);

* research and development (Comin and Philippon, 2005);

* a changing investment opportunity set (Guo and Savickas, 2005);

* growth options (Cao, Simin, and Zhao, 2008); or

* competitive markets (Irvine and Pontiff, 2009).

Bekaert, Hodrick, and Zhang (2008) argue that idiosyncratic volatility is mean reverting and that the factors that many researchers cite as explanations are driven by macroeconomic factors; therefore, idiosyncratic volatility coincides with the business cycle. Brandt et al. (2010) provide evidence that the increase in idiosyncratic volatility may be due to speculative trading, especially in low-priced stocks. Brown and Kapadia (2007) find evidence that there is an increase in the smaller, riskier companies with equity that is publicly traded and that this is driving the observed increase in idiosyncratic volatility, consistent with the observations by Brandt et al. (2010).

Gaspar and Massa (2006) hypothesize that market power reduces idiosyncratic risk through the ability of a firm with such power to pass along cost shocks to customers and because of the reduction in uncertainty regarding profitability. They argue that this may explain the increase in volatility that accompanies deregulation and the globalization of markets. Their evidence indicates that idiosyncratic volatility increases with most cases of deregulation, and that systematic volatility decreases in some cases. (22) Gaspar and Massa (2006) make these comparisons using both 5-year and 12-year averages before and after deregulation.

The results of the Irvine and Pontiff (2009) are consistent with the idea that increased competition is associated with increased idiosyncratic risk. Irvine and Pontiff (2009) examine idiosyncratic risk for seven deregulated industry events: airlines, banks and thrifts, entertainment, natural gas, telecommunication, trucking, and utilities. Irvine and Pontiff (2009) focus on the 480 months from 1964 through 2003, with a resulting difference in lengths of "before" and "after" deregulation periods for these industries.

III. Hypotheses

One possible explanation for changing risk is that competition in the industry has changed, yet it is not clear from both theory and prior evidence whether competition is a systematic risk or an unsystematic risk. If competitive risk is systematic risk, we expect to find an increase in the systematic risk of stocks of deregulated companies following deregulation. If competitive risk is an unsystematic risk of a firm, we expect to find that relative idiosyncratic risk increases following deregulation.

As firms shift from a market with limited competition to a deregulated, more competitive environment, there is a period of time in which firms adjust to the changed environment. This is also a period in which investors are adjusting expectations regarding the firms' ability to adjust and compete in the deregulated market. There is no reason to believe that this adjustment period is the same length for all industries because of the different types of regulation that they face and the different forms that deregulation takes. For example, telecommunication firms still faced some form of regulation in rate setting on the state level after 1996. Consistent with Irvine and Pontiff (2009), as well as Gaspar and Massa (2006), we hypothesize that idiosyncratic risk increases following deregulation. We attribute this to the adjustment to the deregulated market. Therefore, we expect relative idiosyncratic risk to increase following deregulation, but then revert after some time once the firms adjust to the new, more competitive environment. Such a pattern supports the hypothesis that a period of learning follows deregulation.

IV. Data and Methodology

Our investigation focuses on the effect of the deregulation acts on the systematic and unsystematic variance of returns in the deregulated industry. We first identify four different deregulation acts and define the deregulation event date as the day the act passes congress. We provide the acts, enactment date, and two-digit or four-digit standard industrial classification (SIC) codes of the affected firms in Table I. (23) We recognize that the development of legislation spans time prior to the signing of these acts into law, but we use this date to demarcate before and after deregulation. (24)

Our analysis proceeds by examining the variance of returns and its components using the following set of tests performed for the entire sample, which allows firms to enter and exit the industry, and for the sample of firms that existed both before and after deregulation:

* Shifts in beta;

* Changes in the ratio of average systematic variance to average idiosyncratic variance; and

* Changes in the ratio of the industry portfolio's systematic variance to industry-specific variance.

We provide the tests in the corresponding subsections that follow, and then provide the results of these tests in Section V.

A. Measuring Shifts in Beta

We construct a daily time-series on the value-weighted portfolio excess returns ([R.sub.pt]) for each two-digit or four-digit SIC portfolio of securities under study for 4,000 trading days centered on the enactment date:

[R.sub.pt] = [[n.sub.t]summation over (i = 1)][v.sub.it][r.sub.it]/[v.sub.pt] (1)

where [v.sub.it] and [R.sub.it] are the market value and the excess return of the ith stock in the two-digit or four-digit SIC portfolio at time t, respectively, and [v.sub.pt] is the total portfolio value at time t. The total number of stocks in the portfolio at time t is n.

As we mention earlier, we follow two different approaches in constructing the portfolio return series:

* We allow for changes in the portfolio composition when firms enter or leave the industry.

* We allow for changes in the portfolio composition when firms enter or leave the industry but exclude new entrant following the enactment date.

We then estimate the following multivariable equation for each of the nine value-weighted industry portfolios in our study twice. Once for the complete portfolio, and once for the portfolio that excludes the new entrants:

[R.sub.pt] = [[alpha].sub.p] + [[beta].sub.p][R.sub.Mt] + [[lambda].sub.p]([D.sub.p]) + [[beta].sub.ps]([D.sub.p][R.sub.Mt]) + [e.sub.pt], (2)

where [R.sub.Mt] is the Center for Research in Security Prices (CRSP) value-weighted index excess returns at time t, [[lambda].sub.p] is the average daily abnormal return in the period following the enactment date for portfolio p, [D.sub.p] is the postenactment-period event dummy for portfolio p, and [[beta].sub.ps] is the average shift in the portfolio's beta following the enactment date.

We supplement the market model in Equation (2) with a dummy variable that takes on the value 1 in the postenactment period, and zero in the otherwise. This approach is useful in testing the effect of specific events on the beta of stocks. (25) If beta increases following deregulation, we should see this reflected in the estimated [[beta].sub.ps] coefficients.

We collect the stock return data from the University of Chicago's CRSP database and we construct the nine portfolios based on the SIC codes. We provide a description of each of the SIC codes under study in Table II.

B. Estimating Changes in Industry's Average Risk Components

Next, we focus on changes in the average risk components of the industry. We regress the excess return of the ith stock in the two-digit or four-digit SIC portfolio for the trading day t, [R.sub.it], on the CRSP value-weighted index excess returns for that trading day, [R.sub.Mt]:

[R.sub.it] = [[alpha].sub.i] + [[beta].sub.i][R.sub.Mt] + [e.sub.it], (3)

where [[alpha].sub.i] is the intercept, [[beta].sub.i] is stock i's beta, and [e.sub.it] is the stock-specific residual. We separate the variance of each stock's excess returns into market and idiosyncratic components:

[[sigma].sup.2]([R.sub.it]) = [[beta].sup.2.sub.i][[alpha].sup.2]([R.sub.Mt]) + [[sigma].sup.2][e.sub.it]). (4)

We then construct a daily time-series on the stock's beta, as well as its market and idiosyncratic components of the variance using a 300-day moving window. (26) Using these daily series for the individual stocks in each industry, we define the average idiosyncratic variance for each industry portfolio, [IDIO.sub.pt], as

[IDIO.sub.pt] = [[n.sub.t]summation over (i = 1)][v.sub.it][[sigma].sup.2]([e.sub.it])/[v.sub.pt] (5)

and define the average systematic variance for each industry portfolio, [SYS.sub.pt], as

[SYS.sub.pt] = [[n.sub.t]summation over (i = 1)][v.sub.it][[beta].sup.2.sub.it][[sigma].sup.2]([R.sub.Mt])/[v.sub.pt] (6)

Because our investigation focuses on the relation between the market and idiosyncratic variances following deregulation, we examine the 300-day moving window of each portfolio's ratio of average systematic to average idiosyncratic variances ([[delta].sub.pt]):

[[delta].sub.pt] = [SYS.sub.pt]/[IDIO.sub.pt] (7)

We use a dummy-variable event analysis on the moving ratio for each portfolio, and use four dummy variables to examine the changes in the ratio levels under the deregulation regimes:

Regime 1: The first 300 days following the relevant enactment date.

Regime 2: The second 300 days following the relevant enactment date.

Regime 3: The third 300 trading days following the relevant enactment date.

Regime 4: The remaining 1,100 days in the sample, (i.e., the period that starts 901 days following the relevant enactment date.

The evidence of Fama and French (2004) is consistent with the explanation that changes in the idiosyncratic returns volatility could be, in part, due to trends in the entry and exit of firms into the markets. Consistent with this, Wei and Zhang (2006) find that the entrance of newly listed firms is associated with increase in return on equity volatility. If new entrants have greater fundamental volatility and deregulation decreases barriers to entry, the changes in average idiosyncratic risk in an industry following deregulation could be attributed to the changes in the industry composition. We include the Herfindahl Index as a proxy for market concentration in our model to control for changes in market concentration.

[[delta].sub.pt] = [[alpha].sub.p] + [[gamma].sub.p](T) + [[psi].sub.p]([Herf.sub.p]) + [[phi].sub.1p]([D.sub.1p]) + [[phi].sub.2p]([D.sub.2p]) + [[phi].sub.3p]([D.sub.3p]) + [[phi].sub.4p]([D.sub.4p]) + [e.sub.pt]. (8)

The dependent variable is the ratio of the systematic to idiosyncratic variance for the industry, [[delta].sub.pt]. Some researchers report general trends in idiosyncratic risk, so we include T as a proxy for time to account for the possible general trend; therefore, [[gamma].sub.p] captures the temporal trend in [([delta].sub.pt]. (27) We use [Herf.sub.p] as a proxy for market concentration in the industry quarterly Herfindahl Index for portfolio p, so therefore [[psi].sub.p] measures the effect of market concentration on the ratio. We employ dummy variables for regimes 1, 2, 3, and 4, respectively [D.sub.1p], [D.sub.2p], [D.sub.3p], and [D.sub.4p], and the coefficients of these dummy variables, [[phi].sub.1p], [[phi].sub.2p], [[phi].sub.3p], and [[phi].sub.4p], respectively, capture the level change in the ratio under regimes. We therefore use Equation (8) to examine the changes in the levels of the ratio following the enactment date, controlling for the change in the level of competition, portfolio composition, and time.

C. Estimating Changes in Industry Systematic and Industry-Specific Risk

Next, we examine the changes in the components of the variance for the industry as a portfolio. Decomposing the risk of the industry as a portfolio and contrasting the results with those generated using the approach in Section IV. B above, allows us to examine whether changes in the correlation structure between the component firms following deregulation contribute to different dynamics in the risk structure of the industry.

We use the value-weighted return on the industry portfolio, as constructed in Equation (1), and again regress the portfolio excess return for a given trading day ([R.sub.pt]) on the same day CRSP value-weighted index excess returns ([R.sub.Mt]),

[R.sub.pt] = [[alpha].sub.p] + [[beta].sub.p][R.sub.Mt] + [e.sub.pt], (9)

where [[alpha].sub.p] is the intercept, [[beta].sub.p] is the portfolio's beta, and [e.sub.pt] is the portfolio-specific residual. We then decompose the variance of the industry portfolio's excess returns into a systematic component and an industry-specific component:

[[sigma].sup.2][R.sub.pt]) = [[beta].sup.2.sub.p][[sigma].sup.2]([R.sub.Mt]) + [[sigma].sup.2]([e.sub.pt]). (10)

As before, we construct a daily time-series on the portfolios' beta, and the systematic and industry-specific variance using a 300-day moving window.

We define the 300-day moving ratio of systematic to industry-specific variances for each industry portfolio as:

[[delta].sub.It] = [[beta].sup.2.sub.pt][[sigma].sup.2]([R.sub.Mt])/[[sigma].sup.2]([e.sub.pt]), (11)

where [[delta].sub.It] is the daily ratio for industry I at time t, based on the systematic and industry-specific variances of the previous 300 days.

From this point onward, we follow the same analysis introduced in Section IV.B, in which we define the industry portfolio with and without postenactment entrants and evaluate Equation (8) with [delta]It as the dependent variable.

V. Results

We provide the results pertaining to the shifts in beta in Section V.A, the discussion of the results pertaining to the temporal changes in average systematic and average idiosyncratic variances in Section V.B, and the discussion of the results pertaining to the temporal changes in the industry systematic and industry-specific variances in Section VC.

A. Shifts in Beta Following Deregulation

We estimate Equation (4) for each of the nine portfolios in our study and present the results for the complete industry portfolio in Panel A of Table III, and the results for the industry portfolio excluding postenactment entrants in Panel B. We find that the adjustment in the complete industry portfolio's beta is negative and significant in six of the nine portfolios, and positive and significant for insurance brokers only. When we exclude postenactment entrants to the industry, we find that

the adjustment is negative and significant in seven of the nine portfolios, and positive and significant in one, again, the insurance brokers' portfolio. These results are consistent with findings of other researchers who report a general decrease in the market risk of deregulated firms following deregulation. (28) These results do not, however, support the idea that competition is a systematic risk.

[FIGURE 1 OMITTED]

For further illustration, we estimate a 300-day moving window on the beta of each portfolio and present a plot of the resulting-time series in Figure 1. The moving beta's pattern postderegulation confirms the results that we present in Table III.

B. Changes in the Average Components of the Variance Following Deregulation

In line with Irvine and Pontiff (2009), we examine whether the industries experienced changes in the average idiosyncratic volatility following deregulation that are higher or lower than normal. Specifically, we examine the mean values of the ratio of average idiosyncratic to total variance for each industry in the sample, for each of the periods before and after the deregulation act. We use a 600-day range centered on the enactment date and present the results in Table IV, Panel A.

We observe an increase in average relative idiosyncratic risk following deregulation in seven of nine industries, and a decrease in the remaining two. Irvine and Pontiff (2009) interpret the increase in idiosyncratic risk in the deregulated industries as evidence of a decrease in market power because regulated industries have higher barriers to entry that, in turn, enhanced market power. Expanding the estimation range to 4,000 days centered on the enactment date shown in Panel B of Table IV, we find that only three of the nine industries exhibit an increase in relative idiosyncratic risk, with the other six exhibiting a decrease. In other words, there appears to be an increase in idiosyncratic risk following deregulation, but idiosyncratic risk actually decreases once we consider a broader period postderegulation.

Repeating this analysis for each of the industries after excluding postenactment entrants we present the results for the 600 days centered on the enactment in Panel A of Table V and the results for the broader period of 4,000 days centered on enactment in Panel B. We find similar results when excluding postenactment entrants. (29) The bottom line is that idiosyncratic risk generally declines following deregulation, despite an initial increase.

We expand this analysis by examining the mean values of the average idiosyncratic risk before deregulation and across the four different regimes identified earlier. We present the results in Panel A of Table VI. We find that relative idiosyncratic risk decreased across all four regimes, on average, when compared with the prederegulation period. The lowest average decrease occurs in the first regime followed by the second regime, whereas the highest average decrease occurs in the third regime and then drops again in the fourth regime. We find the same pattern across the four regimes when we exclude postenactment entrants from the industry average idiosyncratic risk, shown in Panel B of Table IV.

C. Detecting a Pattern of Adjustment in Average Risk Components

Next, we use the ratio of average systematic to average idiosyncratic variance ([[delta].sub.pt]) to examine the possibility of shifting systematic risks that is not sensitive to a general increase in total risk. We provide descriptive statistics on the ratio ([[delta].sub.pt]) for the complete industry and for the industry excluding new entrants following the enactment date are presented in Table VII in Panels A and B, respectively. (30)

We also account for the possible decrease in market power as hypothesized in Irvine and Pontiff (2009) by including the Herfindahl Index as a proxy for market concentration, as we present in Equation (8). We estimate Equation (8) for each of the portfolios in our study and present the results in Table VIII.

A higher value for the Herfindahl Index implies a less competitive, and more concentrated, industry; hence, a positive [psi] in Equation (8) indicates that a less competitive industry is associated with a higher ratio of average systematic to average idiosyncratic variance or risk; a negative [psi] indicates that a less competitive industry is associated with a lower ratio of average systematic to average idiosyncratic risk.

We find that the estimated coefficient associated with the Herfindahl Index ([psi]) is positive and significant in six of the nine portfolios, negative and significant in one, and insignificant in the remaining two portfolios. In other words, in most cases, as the industry became less concentrated and more competitive, its ratio of average systematic to average idiosyncratic risk increased.

The estimated coefficients on the regime dummy variables in Equation (8) provide information on whether the ratio of average systematic to average idiosyncratic risk increased or decreased relative to the prederegulation period. In four of the nine industries under study, the ratio of average systematic to average idiosyncratic variance shows a significant decrease in the first regime following the enactment date compared to the preenactment period. The ratio increased in two of the nine industries. In the second regime, we find similar overall results.

Examining the first and second regimes together, we find that five of the nine industries exhibit a decrease in the ratio of average systematic to average idiosyncratic variance, two industries exhibit an increase in the ratio, and two do not exhibit any significant change. By the third regime, however, we find that eight of the nine industries exhibit an increase in their ratios compared to the preenactment period, the only exception being the air transportation industry, which had notably exhibited an increase in the ratio in the first regime.

In the fourth regime, we find that only three industries exhibit an increase in the ratio compared to the preenactment period, two industries exhibit a decrease and then insignificant values for the remaining industries. In other words, by the fourth regime most industries had returned to ratio levels that are comparable with the preenactment period.

The only industry to exhibit an increase in the ratio throughout the four postenactment regimes is the communication industry. Notably, this industry is the only one that exhibited a negative temporal trend in its ratio throughout the 4,000-day period. We find similar results when we repeat this analysis for each industry excluding postenactment entrants. We present the results in Table IX.

We plot the time series of the industries' ratios throughout the study period and note the four regimes in Figure 2. The initial dips in the ratios across regimes 1 and 2, the increase in regime 3, and the ensuing decrease in regime 4 are clearly visible across all industries except airlines and telecommunication. For a clearer illustration of the changes in the level of the ratio across the nine industries, we plot the average levels of the detrended ratios across the four regimes based on the results presented in Table VIII and present the graph in Figure 3. We also present separate graphs for the ratio of each of the industries along with the systematic and idiosyncratic proportions from the total variance in Figure 4.

As apparent in Table VIII and in Figures 2 through 4, the results are consistent with our hypothesis of an episodic increase in the idiosyncratic variance relative to the systematic variance following deregulation even after accounting for market concentration. A possible explanation is that this episodic increase represents learning; that is, it takes time for firms within an industry to learn how to deal with competition in a deregulated market and for investors in that market to realign their expectations. Relative idiosyncratic risk diminishes as firms and investors learn.

[FIGURE 3 OMITTED]

D. Changes in the Components of the Industry Portfolio Variance

We repeat the analysis presented in Section V.B for the industry as a portfolio rather than for the average firm in the industry. This analysis allows us to examine whether changes in the correlation structure between firms following deregulation might cause different dynamics in the risk adjustment for the industry as a portfolio rather than for the average firm in the industry. We begin by examining the mean values of the ratio of the industry-specific variance to total variance for each industry in the sample, for each of the periods before and after the deregulation act. We use a 600-day range centered on the enactment date and present the results in Panel A of Table X.

[FIGURE 4 OMITTED]

We observe an increase in relative industry-specific variance following deregulation in eight of the nine industry portfolios and a decrease in the remaining one. We then expand the estimation range to 4,000 days centered on the enactment date and present the results in Panel B of Table X. We find that only four of the nine industries exhibit an increase in idiosyncratic risk and the other five industries exhibit a decrease. We find similar results when we repeat this analysis after excluding postenactment entrants from the industry portfolios. We present the results in Table XI.

Overall these results suggest that, similar to the increases in average idiosyncratic risk, increases in industry-specific risk following deregulation are temporary and diminish with a longer analysis horizon.

We expand this analysis by examining the mean values of the industry-specific risk before deregulation and across the four different regimes identified earlier. We present the results in Panel A of Table XII. The increases in industry-specific risk occur in the first and second regime, that is, in the first 600 days following the enactment. We observe an increase in industry-specific risk for seven of the nine portfolios in the first regime but only for five in the second regime as compared to the prederegulation period.

This pattern changes dramatically in the third regime, in which six of the nine portfolios exhibit a drop in their industry-specific risk compared to the preenactment period. The average change in industry-specific risk in the third regime is negative as compared to the preenactment period. By the fourth regime, eight of the nine portfolios exhibit a decrease in their relative industry-specific risk as compared with the preenactment period. We detect an almost identical pattern when we exclude postenactment entrants from the portfolios, as we show in Panel B of Table XII.

Next, we use the ratio of industry systematic to industry-specific variance ([[delta].sub.It]) to examine the shifting risks in the portfolio in a manner that is not sensitive to a general increase in its total risk. Similar to our approach in Section VB, we account for the possible decrease in market power by including the Herfindahl Index as a proxy for market concentration. We estimate Equation 8 for each of the portfolios in our study and present the results in Table XIII.

In six of the nine industry portfolios, the ratio of industry systematic to industry-specific variances shows a significant decrease in the first regime following the enactment date compared to the preenactment period. The ratio increased in two of the nine regimes. In other words, in most cases industry-specific variance increased relative to industry systematic variance in the 300-day period following the deregulation.

Of the six industries that had a significant decrease in their ratio in the first regime, four remained as such in the second regime. We find that the ratio for the security and commodity brokers industry shows a reversal to a significant increase throughout the second, third, and fourth regimes, as compared to the preenactment period. In other words, the security and commodity brokers industry and bank holding companies experienced only a very temporary increase in industry-specific risk, relative to industry systematic risk. A possible explanation is that if this ratio is a proxy for learning, this industry learned faster than the others.

[FIGURE 5 OMITTED]

In the third regime, seven of the nine industries, with the exception of nondepository credit institutions and air transportation, exhibited a reversal to either an increase in the ratio level as compared to the preenactment period or no significant difference in the ratio compared to the preenactment period. In the fourth regime, only one industry portfolio, nondepository credit institutions, had a decrease in the ratio compared to the preenactment period, albeit a much lesser decrease compared to regimes 1 and 2.

The only portfolio to exhibit a positive increase in the ratio of systematic to industry-specific risk throughout the four postenactment regimes is, again, the communication industry. This portfolio is also the only one that exhibited a negative and significant temporal trend in its ratio throughout the 4,000-day period.

[FIGURE 6 OMITTED]

We plot the time series of the industry portfolios' ratios throughout the study period and note the four regimes in Figure 5. The initial dip in the ratios across regimes 1 and 2 and the ensuing increase are clearly visible across all industries except the airline and communication industries. For a clearer illustration of the changes in the level of the ratio across the nine portfolios, we plot the average levels of the detrended ratios across the four regimes based on the results presented in Table XIII and present the graph in Panel A of Figure 5. We also present separate graphs for the ratio of each of the portfolios along with the industry systematic and industry-specific proportions from the total variance in Figure 7.

[FIGURE 7 OMITTED]

Examining the ratio after excluding postenactment entrants is particularly important in this section because a possible interpretation for the perceived decrease in the relative industry-specific variance (that is, an increase in the ratio) in later regimes might be due simply to the effect of diversification on the portfolio. Deregulation reduces barriers to entry allowing for more firms to join the industry. As more firms enter the industry portfolio, the diversification effect is expected to increase, thus leading to a possible reduction in the relative industry-specific risk of the whole portfolio.

We reevaluate Equation (8) for each of the reconstructed portfolios and present the results in Table XIV. We see the same pattern identified in the portfolios of all firms (from Table XIII) holds in the portfolios excluding postenactment entrants. The only exception is the portfolio of bank holding companies, where we do not find the initial dip in the ratio in regime 1 to be significant and the trucking portfolio where we find the initial dip in the ratio in regime 1 to be significant.

For a clearer illustration of the changes in the level of the ratio across the nine portfolios after excluding the postenactment entrants, we plot the average levels of the detrended ratios across the four regimes based on the results presented in Table XIV and present the graph in Panel B of Figure 6.

As apparent in Tables XIII and XIV and in Figures 5 through 7, the results are, again, consistent with an episodic increase in the industry-specific variance relative to systematic variance following deregulation even after accounting for market concentration.

VI. Conclusion

We examine the market's perception of risk associated with deregulation in this study. We decompose security risk for nine industries around deregulation into idiosyncratic and systematic risk. We find a consistent pattern of risk adjustment to deregulation: increased security risk is temporary and largely idiosyncratic. These results are consistent with the idea that firms in deregulated industries learn to deal with competition.

Our findings reconcile the results of other studies that examine risk change around deregulation. If researchers focus only on a segment of trading days immediately following deregulation, a reasonable conclusion is that risk increased following deregulation. However, extending the analysis further and decomposing risk into systematic and unsystematic variances reveals a pattern that is more sinusoidal in nature.

Our examination of the ratio of systematic to unsystematic risk reveals that there is a pattern of shifting risk from systematic to unsystematic immediately following deregulation, but that within three years both systematic and unsystematic risk decline relative to prederegulation levels. We suggest that firms learn to compete following deregulation and that this learning, though different for each industry, may span a period of three years; during this learning period, the firms' idiosyncratic risk increases relative to the firms' systematic risk.

The implications of these findings are threefold. First, simply examining a firm's systematic risk and comparing it before and after deregulation does not provide a complete picture of the changes in firms' risk in response to deregulations. Second, it takes time to learn to compete in the deregulated environment and therefore an assessment of competitive effects, especially with respect to risk, must be viewed over a period extending beyond three years. Third, in most cases in our analysis, deregulation results in only temporary changes in risk level and composition.

We thank the two anonymous referees for their helpful comments. The suggestions resulted in a significantly improved paper.

References

Aharony, J., A. Saunders, and I. Swary, 1988, "The Effects of DIDMCA on Bank Stockholders' Returns and Risk," Journal of Banking and Finance 12, 317-331.

Akhigbe, A. and A.M. Whyte 2001, "The Impact of FDICIA on Bank Returns and Risk: Evidence for the Capital Markets," Journal of Banking and Finance 25, 393-417.

Allen, P.R. and W.J. Wilhelm, 1988, "The Impact of the 1980 Depository Institutions Deregulation and Monetary Control Act on Market Value and Risk: Evidence from the Capital Markets," Journal of Money, Credit and Banking 20, 364-380.

Allen, S.A., L.F. Cunningham, and W.R. Wood, 1990, "Airline Industry Deregulation and Changes in Systematic Risk," Transportation 17, 49-66.

Babcock, M.W., 1984, "Efficiency and Adjustment: The Impact of Railroad Deregulation," Cato Institute, Cato Policy Analysis, No. 33, pp. 1-13.

Bakaert, G., R.J. Hodrick, and X. Zhang, 2008, "Is There a Trend in Idiosyncratic Volatility?" Columbia University Working paper.

Bali, T., N. Cakici, and H. Levy, 2008, "A Model-Independent Measure of Aggregate Idiosyncratic Risk," Journal of Empirical Finance 15, 878-896.

Bali, T., N. Cakici, X. Yah, and Z. Zhang, 2005, "Does Idiosyncratic Risk Really Matter?" Journal of Finance 60, 905-929.

Banker, R.D., S. Das, and C.S. Ou, 1997, "Shareholder Wealth Effects of Legislative Events: The Case of Airline Deregulation," Public Choice 91, 301-331.

Beneish, M.D., 1991, "The Effect of Regulatory Changes in the Airline Industry on Shareholders' Wealth," Journal of Law and Economics 34, 395-430.

Brandt, M.W., A. Bray, J.R. Graham, and A. Kumar, 2010, "The Idiosyncratic Volatility Puzzle: Time Trend or Speculative Episodes?" Review of Financial Studies 23, 863-899.

Brown, G. and N. Kapadia, 2007, "Firm-Specific Risk and Equity Market Development," Journal of Financial Economics 84, 358-388.

Campbell, J., M. Lettau, B.G. Malkiel, and Y. Xu, 2001, "Have Individual Stocks Become More Volatile? An Empirical Exploration of Idiosyncratic Risk," Journal of Finance 56, 1-43.

Cao, C., T. Simin, and J. Zhao, 2008, "Can Growth Options Explain the Trend in Idiosyncratic Risk?" Review of Financial Studies 21, 2599-2633.

Carow, K.A. and R.A. Heron, 2002. "Capital Market Reactions to the Passage of the Financial Services Modernization Act of 1999," Quarterly Review of Economics and Finance 42, 465-485.

Cavarra, L., R.D. Stover, and B.J. Allen, 1981, "The Capital Market Effects of Airline Deregulation," Transportation Journal 20, 73-78.

Chen, A. and G. Sanger, 1985, "An Analysis of the Impact of Regulatory Change: The Case of Natural Gas Deregulation," Financial Review 20, 36-54.

Comin, D. and T. Philippon, 2005, "The Rise in Firm-Level Volatility: Causes and Consequences," NBER Macroannuals.

Correa, R. and G. Suarez, 2009, "Firm Volatility and Banks: Evidence from U.S. Banking Deregulation," Federal Reserve Board, Finance and Economics Discussion Series.

Cunningham, L.E, M.B. Slovin, W.R. Wood, and J.K. Zaima, 1988, "Systematic Risk in the Deregulated Airline Industry," Journal of Transport Economics and Policy 22, 345-353.

Cyree, K.B., 2000, "The Erosion of the Glass-Steagall Act: Winners and Losers in the Banking Industry," Journal of Economies and Business 52, 343-363.

Ely, D.R and K.J. Robinson, 1999, "The Determinants of the Wealth Effects of Banks' Expanded Securities Powers," Working paper, Federal Reserve Bank of Dallas Financial Industry Studies.

Enis, C.R. and E.A. Morash, 1984, "Some Aspects of Motor Carrier Size, Concentration Tendencies, and Performance after Deregulation," Akron Business and Economic Review 18, 82-94.

Enis, C.R. and E.A. Morash, 1985, "Accounting for Public Policy Actions: The Case of Motor Carrier Deregulation," ABACUS 21, 63-83.

Fama, E. and K. French, 2004, "New Lists and Seasoned Firms: Fundamentals and Survival Rates," Journal of Financial Economics 73, 229-269.

Fink, J., K. Fink, G. Grullon, and J. Weston, 2005, "IPO Vintage and the Rise of Idiosyncratic Risk," Rice University Working paper.

Fraser, D.R. and S. Kannan, 1990, "Deregulation and Risk: Evidence from Earnings Forecasts and Stock Prices," Financial Management 19, 68-76.

Gaspar, J. and M. Massa, 2006, "Idiosyncratic Volatility and Product Market Competition," Journal of Business 79, 3125-3152.

Guo, H. and R. Savickas, 2005, "Average Idiosyncratic Volatility in G7 Countries," Review of Financial Studies 21, 1259-1296.

Guthrie, G., 2006, "Regulating Infrastructure: The Impact on Risk and Investment," Journal of Economic Literature 44, 925-972.

Horwitz, R.B., 1991, The Irony of Regulatory Reform, New York, Oxford University Press.

Irvine, P.J. and J. Pontiff, 2009, "Idiosyncratic Return Volatility, Cash Flows, and Product Market Competition," Review of Financial Studies 22, 1149-1177.

Joskow, P.L. and P. MacAvoy, 1975, "Regulation and the Financial Condition of Electric Power Companies in the 1970s," American Economic Review 65, 36-54.

Joskow, P.L., 1974, "Inflation and Environmental Concern: Structural Change in the Process of Public Utility Price Regulation," Journal of Law and Economics 17, 291-327.

Mamun, A., M.K. Hassan, G.V. Karels, and N. Maroney, 2005, "Financial Services Modernization Act of 1999: Market Assessment of Winners and Losers in the Insurance Industry," Journal of Insurance Issues 28, 103-128.

Morash, E.A., E. Bruning, and D.L. McQuin, 1981, "Motor Carrier Capital Costs and Deregulation: A Tentative Assessment," Transportation Research Forum 22, 449-466.

Neale, F.R., P.P. Drake, and S. Clark, 2010, "The Financial Services Industry and the Modernization Legislation," B. E. Journal of Economic Analysis and Policy, 10(1), Article 16.

Neale, F.R. and P.P. Peterson, 2005, "The Effect of the Gramm-Leach-Bliley Act on the Insurance Industry," Journal of Economics and Business 57, 317-338.

O'Connor, W.E., 1995, An Introduction to Airline Economics, 5th Ed., Westport, CT, Praeger.

Patashnik, E.M., 2008, Reforms at Risk: What Happens after Major Policy Changes Are Enacted, Princeton, NJ, Princeton University Press.

Peltzman, S., 1976, "Toward a More General Theory of Regulation," Journal of Law and Economics 19, 211-240.

Rose, M.H., B.E. Seely, and P.F. Barrett, 2006, The Best Transportation System in the World, Columbus, Ohio State University Press.

Saunders, A. and M. Smirlock, 1987, "Intra- and Interindustry Effects of Bank Securities Market Activities: The Case of Discount Brokerage," Journal of Financial and Quantitative Analysis 22, 467-482.

Strahan, P.E. and A. Sufi, 2001, "Who Gains the Most?" Federal Reserve Bank of Chicago Bank Structure Conference Proceedings, 682-698.

Van Auken, H. and M.R. Crum, 1985, "The Impact of the Motor Carrier Act of 1980 on Motor Carrier Common Stock Returns," Transportation Research 19A, 43-49.

Wei, S. and C. Zhang, 2005, "Idiosyncratic Risk Does Not Matter: A Re-examination of the Relationship between Average Returns and Average Volatilities," Journal of Banking and Finance 29, 603-621.

Wei, S. and C. Zhang, 2006, "Why Did Individual Stocks Become More Volatile?" Journal of Business 79, 259-292.

Yildirim, H.S., S.A. Kwag, and M.C. Collins, 2006, "An Examination of the Equity Market Response to the Gramm-Leach-Bliley Act across Commercial Banking, Investment Banking and Insurance Firms," Journal of Business Finance & Accounting 33, 1629-1649.

(1) See Peltzman (1976).

(2) Following an episode of excessive competition, the Civil Aeronautics Board in 1937 allowed airlines to establish a cartel, with the purpose of providing stability to the then fledgling industry. Essentially, market shares were divided among the airlines until the deregulation opened up competition.

(3) See Joskow and MacAvoy (1975) and Keran (1976). Joskow (1974) argues that rate decreases under rate regulation were quickly enacted, but rate increases were time consuming to enact.

(4) For example, it is estimated that it took the Civil Aeronautics Board up to 4.5 years to permit airline fare increases in the 1960s (Horowitz, 1991).

(5) The relationship between regulation and investments is discussed in Guthrie (2006).

(6) Patashnik (2008).

(7) Public Law 95-504.

(8) The entry into the market and fare-setting ability were phased in during 1981 and 1982, respectively.

(9) At the time of deregulation, there were six major airlines. By 1991, there were three of these remaining, with a large number of smaller airlines (O'Connor, 1995).

(10) We can trace trucking deregulation to the early 1970s, but the 1978 decision by the Interstate Commerce Commission (ICC) that allowed private firms, such as manufacturers, to carry loads of others to fill their trucks, was the first significant act that went against the previous practice (Toto Purchasing and Supply Company, Inc. decision). Starting in October 1979, the ICC began to loosen entry into the industry (Rose, Seely, and Barrett, 2006). Public Law S.2245, also known as the Motor Carrier Act of 1980, was signed in to law July l, 1980.

(11) This accounting standard, FASB 44, was effective for fiscal years ending after December 15, 1980, which affects the fiscal year of the MCA. Enis and Morash (1985) conclude that there was no effect of the write-off on the trucking company shareholder wealth.

(12) Title 47 U.S.C. 151.

(13) The case of United States" v. AT&T was settled on January 8, 1982.

(14) TERMCA, S. 173, was signed into law November 22, 1991. The Telecommunications Act was passed February 1, 1996 and signed into law February 9, 1996 [Public Law 104-104].

(15) Not only did this act address opening competition, it also had provisions regarding obscenity and violence in programming and unfair billing practices.

(16) US Public Law No. 106-102, signed into law November 12, 1999.

(17) The Banking Act of 1933 is commonly known as the Glass-Steagall Act [June 16, 1933, Ch 89 [section] 20, 48 Stat. 188; Aug 23, 1935, Ch 614, Title III, [section] 302, 49 Star. 707]. The Bank Holding Act of 1956 [12 U. S. C. [section] 1841 ] allowed establishment of bank holding companies, but prohibited a bank holding company in one state from acquiring a bank in another state.

(18) Mamun, Hassan, and Lai (2004) also find that systematic risk was reduced for banks, insurers, and brokerage firms after passage of the FSM Act.

(19) Public Law 94-210, February 5, 1976.

(20) Public Law 96-448, October 14, 1980.

(21) Babcock (1984).

(22) Gaspar and Massa (2006) examine the deregulated industries of airlines, electricity, natural gas, telecommunications (AT&T break-up only), and transportation.

(23) Though some researchers criticize the use of the SIC code to define an industry, in the case of deregulated industries the industry is well defined by the regulation, and therefore there is less risk using this code classification.

(24) Because there is likely some anticipation of deregulation, the signing of the act in to law represents the point in time when the changes are effective and the most significant deregulation takes place.

(25) See, for example, Allen, Cunningham, and Wood (1990), Ely and Robinson (1999), Cyree (2000), Carow and Heron (2002), and Neale and Peterson (2005).

(26) We use two alternative approaches to calculating the total variance and its components for subsections 4.B and 4.C using non-overlapping monthly observations, where the variance of each month is calculated based on the daily returns throughout that month similar to Campbell et al. (2001) and Irvine and Pontiff (2009). We repeat the same approach with quarterly instead of monthly variances. The results from these alternative measurements of variance remained consistent with our findings in Section V. The results are available upon request.

(27) See, for example, Campbell et al. 2001), Fama and French (2004), Fink et al. (2005), Wei and Zhang (2006), Bali, Cakici, and Levy (2008), Brandt et al. (2010) and Bakaert, Hodrick, and Zhang (2008).

(28) See, for example, Allen, Cunningham, and Wood (1990), Beneish (1991), Mamun et al. (2004), Neale and Peterson (2005), and Neale, Drake, and Clark (2010).

(29) When we expand the estimation range, the trucking industry exhibits a decrease in average idiosyncratic risk instead of an increase.

(30) For further illustration, we plot [[delta].sub.pt] for each industry across the entire period in Figure 2. We discuss this figure further on.

Elias Semaan and Pamela Peterson Drake *

* Elias Semaan is an Assistant Professor of Finance and Pamela Peterson Drake is a Professor of Finance, both at James Madison University, Harrisonburg, Virginia 22807.
Table I. Deregulation Acts, Enactment Dates, and the Two-Digit or
Four-Digit SIC Codes of Affected Companies

       Deregulation Act             Enactment          SIC Codes of
                                       Date         Affected Companies

Airline Deregulation Act         February 6, 1978          4512
Motor Carrier Regulatory         July 1, 1980              4213
  Reform and Modernization Act
Telecommunications Act of 1996   January 3, 1996           4813
Gramm-Leach-Bliley Financial     November 4, 1999    60, 61, 62, 63,
  Services Modernization Act                              64, 6712

Table II. Description of the SIC Codes

SIC Code                         Description

4213       Trucking, except local
4512       Telephone communications, except radiotelephone
             (communication)
4813       Air transportation, scheduled
60         Depository credit Institutions
61         Nondepository credit institutions
62         Security and commodity brokers, dealers, exchanges, and
             services
63         Insurance carriers
64         Insurance agents, brokers, and services
6712       Bank holding companies

Table III. The Effects of the Deregulation Acts on the Betas of
the Deregulated Industries  We estimate Equation (4) for each of
the nine value-weighted industry portfolios:

[R.sub.pt] = [[alpha].sub.p] + [[beta].sub.p][R.sub.Mt], +
[[lambda].sub.p]([D.sub.p]) + [[beta].sub.ps]([D.sub.p][R.sub.Mt])
+ [e.sub.pt],

where [[alpha].sub.p] is the intercept, [[beta].sub.p] is the
portfolio's [[lambda].sub.p] is the average daily abnormal
return in the period following the enactment date for portfolio
p, [D.sub.p] is the postenactment-period event dummy for
portfolio p, and [[beta].sub.ps] is the average shift in
portfolio p's beta following the enactment date. In Panel A we
report the results  for the portfolios allowing for changes when
firms enter or leave the industry, and in Panel B we report the
results for the portfolios excluding postenactment  entrants. We
report p-values based on the Newey-West standard errors in
parentheses below the coefficient estimates.

Estimated                     Portfolio
Parameter
                 Trucking   Air              Depository
                            Transportation   Communication

  Panel A. Industry including all postenactment entrants

[alpha]          0.00034    0.00029          0.00031
                 (0.180)    (0.445)          (0.103)

[beta]           0.88412    1.70070          0.93000
                 (0.000)    (0.000)          (0.000)

[lambda]         0.00026    -0.00048         -0.00002
                 (0.457)    (0.367)          (0.953)

[[beta].sub.s]   -0.24354   -0.20478         -0.00765
                 (0.000)    (0.002)          (0.796)

[R.sup.2]        0.3136     0.5546           0.3777

  Panel B. Industry  Excluding all Postenactment Entrants

[alpha]          0.00034    0.00029          0.00031
                 (0.221)    (0.485)          (0.015)

[beta]           0.88412    1.70070          0.93000
                 (0.000)    (0.000)          (0.000)

[lambda]         0.00022    -0.00048         -0.00009
                 (0.563)    (0.398)          (0.765)

[[beta].sub.s]   -0.23020   -0.19768         -0.05991
                 (0.000)    (0.011)          (0.046)

[R.sup.2]        0.3088     0.3757           0.4967

Estimated                         Portfolio
Parameter
                 Institutions   Nondepository   Security and
                                Credit          Commodity
                                Institutions    Brokers

Panel A. Industry including all postenactment entrants

[alpha]          0.00055        0.00046         0.00054
                 (0.000)        (0.030)         (0.034

[beta]           1.01051        1.18521         1.47189
                 (0.000)        (0.000)         (0.000)

[lambda]         -0.00007       -0.00003        0.00013
                 (0.759)        (0.908)         (0.718)

[[beta].sub.s]   -0.03442       -0.29196        0.00670
                 (0.098)        (0.000)         (0.857)

[R.sup.2]        0.6815         0.5351          0.6253

  Panel B. Industry  Excluding all Postenactment Entrants

[alpha]          0.00055        0.00046         0.00054
                 (0.000)        (0.019)         (0.030)

[beta]           1.01052        1.18521         1.47189
                 (0.000)        (0.000)         (0.000

[lambda]         -0.00010       -0.00007        0.00009
                 (0.668)        (0.808)         (0.807)

[[beta].sub.s]   -0.03073       -0.29144        0.01505
                 (0.057)        (0.000)         (0.778)

[R.sup.2]        0.6748         0.5301          0.6239

Estimated                    Portfolio
Parameter
                 Insurance   Insurance   Bank
                 Carriers    Brokers     Holding
                                         Companies
  Panel A. Industry including all postenactment entrants

[alpha]          0.00046     0.00031     0.00037
                 (0.002)     (0.150)     (0.000)

[beta]           0.79002     0.75782     0.70939
                 (0.000)     (0.000)     (0.000)

[lambda]         0.00010     0.00032     0.00024
                 (0.639)     (0.293)     (0.049)

[[beta].sub.s]   -0.04965    0.15614     -0.05342
                 (0.019)     (0.000)     (0.000)

[R.sup.2]        0.5784      0.4396      0.7544

  Panel B. Industry  Excluding all Postenactment Entrants

[alpha]          0.00046     0.00031     0.00037
                 (0.000)     (0.085)     (0.000)

[beta]           0.79002     0.75782     0.70939
                 (0.000)     (0.000)     (0.000)

[lambda]         0.00003     0.00007     0.00026
                 (0.878)     (0.842)     (0.054)

[[beta].sub.s]   -0.04872    0.19682     -0.04363
                 (0.037)     (0.000)     (0.014)

[R.sup.2]        0.5719      0.4335      0.7371

Table IV. Pre- and Postderegulation Means of Average
Idiosyncratic Variance Divided by Average Variance per Industry

Table values are the mean values of the average idiosyncratic
variance divided by the average total variance for the pre- and
postderegulation periods for each industry. We use Newey-West
standard errors to determine the significance of the difference
in the means for each industry and then present the average
difference across industries. We estimate this equation over 600
days (Panel A) and 4,000 days (Panel B) centered on the
enactment date of the relevant deregulation act.

Industry                Preenactment     Postenactment     Difference

                       Mean       N      Mean       N

          Panel A. Results Estimated Over 600 Days Centered Around
                            the Enactment Date

Trucking              0.89893    300    0.95037    300     0.05145 ***
Air transportation    0.81193    300    0.68552    300    -0.12641 ***
Communication         0.83650    300    0.82256    300    -0.01394
Depository            0.62179    300    0.71472    300     0.09293 ***
  institutions
Nondepository         0.64363    300    0.74861    300     0.10498 ***
  credit
  institutions
Security and          0.64975    300    0.70235    300     0.05261 ***
  commodity brokers
Insurance carriers    0.72517    300    0.82403    300     0.09886 ***
Insurance brokers     0.69522    300    0.78285    300     0.08763 ***
Bank holding          0.73433    300    0.82734    300     0.09301
  companies

Average difference                                         0.04901 *

     Panel B. Results Estimated Over 4,000 Days Centered Around the
                            Enactment Date

Trucking              0.87941   2,000   0.89052   2,000    0.01111 *
Air transportation    0.74028   2,000   0.76109   2,000    0.02081 ***
Communication         0.68415   2,000   0.79123   2,000    0.10708 ***
Depository            0.79717   2,000   0.62271   2,000   -0.17446 ***
  institutions
Nondepository         0.76481   2,000   0.73466   2,000   -0.03015 ***
  credit
  institutions
Security and          0.77342   2,000   0.57212   2,000   -0.20130 ***
  commodity brokers
Insurance carriers    0.84713   2,000   0.76804   2,000   -0.07909 ***
Insurance brokers     0.86228   2,000   0.70441   2,000   -0.15787 ***
Bank holding          0.83760   2,000   0.71107   2,000   -0.12653 ***
  companies
Average difference                                        -0.07004 *

*** Significant at the 0.01 level.

 * Significant at the 0.10 level.

Table V. Pre- and Post deregulation Means of Average
Idiosyncratic Variance  Divided by Average Variance per Industry,
Excluding Postenactment Entrants

Table values are the mean values of the average idiosyncratic
variance divided by the average total variance for the pre- and
postderegulation periods for each industry excluding
postenactment entrants. We use Newey-West standard errors to
determine the significance of the difference in the means for
each industry and then  present the average difference across
industries. We estimate this equation over 600 days (Panel A) and
4,000 days (Panel B) centered on the enactment date of the
relevant deregulation act.

Industry               Preenactment      Postenactment     Difference

                       Mean       N      Mean       N

     Panel A. Results Estimated Over 600 Days Centered Around the
                           Enactment Date

Trucking              0.89893   300     0.95044   300      0.05151 ***
Air transportation    0.81193   300     0.68552   300     -0.12641 ***
Communication         0.83650   300     0.82099   300     -0.01551 ***
Depository            0.62179   300     0.71383   300      0.09204 ***
  institutions
Nondepository         0.64363   300     0.74682   300      0.10319 ***
  credit
  institutions
Security and          0.64975   300     0.70180   300      0.05205 ***
  commodity brokers
Insurance carriers    0.72517   300     0.82341   300      0.09823 ***
Insurance brokers     0.69522   300     0.78267   300      0.08745 ***
Bank holding          0.73433   300     0.83496   300      0.10063 ***
  companies
Average difference                                         0.04924 *

      Panel B. Results Estimated Over 4,000 Days Centered Around
                          the Enactment Date

Trucking              0.87941   2,000   0.87773   2,000   -0.00168 *
Air transportation    0.74028   2,000   0.75721   2,000    0.01692 **
Communication         0.68415   2,000   0.77963   2,000    0.09548 ***
Depository            0.79717   2,000   0.61522   2,000   -0.18195 ***
  institutions
Nondepository         0.76481   2,000   0.72035   2,000   -0.04446 ***
  credit
  institutions
Security and          0.77342   2,000   0.55333   2,000   -0.22009 ***
  commodity brokers
Insurance carriers    0.84713   2,000   0.75750   2,000   -0.08963 ***
Insurance brokers     0.86228   2,000   0.69765   2,000   -0.16463 ***
Bank holding          0.83760   2,000   0.74151   2,000   -0.09609 ***
  companies
Average difference                                        -0.07624 **

*** Significant at the 0.01 level.
 ** Significant at the 0.05 level.
  * Significant at the 0.10 level.

Table VI. Means of Average Idiosyncratic Variance Divided by
Average Total Variance per Industry, Prior to Deregulation  and
Under the Four Regimes Following Deregulation

Table values are the mean value of the average idiosyncratic
variance divided by average total variance for the
prederegulation period and across each of the  four regimes
following deregulation for each industry. We use Newey-West
standard errors to determine the significance of the difference
in the means between  the respective regime and the
prederegulation period. We also present the average difference
across industries for each regime.

Industry                            Preenactment

                                        Mean

Panel A. Industry Including All Postenactment Entrants

Trucking                              0.87941
Air transportation                    0.74028
Communication                         0.68415
Depository institutions               0.79717
Nondepository credit institutions     0.76481
Security and commodity brokers        0.77342
Insurance carriers                    0.84713
Insurance brokers                     0.86228
Bank holding companies                0.83760
Average difference

Panel B. Industry Excluding All Postenactment Entrants

Trucking                              0.87941
Air transportation                    0.74028
Communication                         0.68415
Depository institutions               0.79717
Nondepository credit institutions     0.76481
Security and commodity brokers        0.77342
Insurance carriers                    0.84713
Insurance brokers                     0.86228
Bank holding companies                0.83760
Average difference

Industry                                   Regime 1

                                     Mean     Difference

Panel A. Industry Including All Postenactment Entrants

Trucking                            0.95036    0.0709 ***
Air transportation                  0.68561   -0.0547 ***
Communication                       0.82253    0.1384 ***
Depository institutions             0.71460   -0.0826 ***
Nondepository credit institutions   0.74838   -0.0164 **
Security and commodity brokers      0.70249   -0.0709 ***
Insurance carriers                  0.82384   -0.0233 ***
Insurance brokers                   0.78282   -0.0795 ***
Bank holding companies              0.82737   -0.01023
Average difference                            -0.01425

Panel B. Industry Excluding All Postenactment Entrants

Trucking                            0.95042    0.0710 ***
Air transportation                  0.68561   -0.0547 ***
Communication                       0.82097    0.1368 ***
Depository institutions             0.71371   -0.0835 ***
Nondepository credit institutions   0.74660   -0.0182 **
Security and commodity brokers      0.70195   -0.0715 ***
Insurance carriers                  0.82322   -0.0239 ***
Insurance brokers                   0.78264   -0.0796 ***
Bank holding companies              0.83495   -0.00265
Average difference                            -0.01402

Industry                                   Regime 2

                                     Mean     Difference

Panel A. Industry Including All Postenactment Entrants

Trucking                            0.95809    0.0787 ***
Air transportation                  0.76776    0.0275 **
Communication                       0.80018    0.1160 ***
Depository institutions             0.67583   -0.1213 ***
Nondepository credit institutions   0.80368    0.0389 ***
Security and commodity brokers      0.55223   -0.2212 ***
Insurance carriers                  0.87132    0.0242 **
Insurance brokers                   0.76624   -0.0960 ***
Bank holding companies              0.75082   -0.0868 ***
Average difference                            -0.02668

Panel B. Industry Excluding All Postenactment Entrants

Trucking                            0.95975    0.0803 ***
Air transportation                  0.76776    0.0275 ***
Communication                       0.79444    0.1103 ***
Depository institutions             0.67283   -0.1243 ***
Nondepository credit institutions   0.79990    0.0351 ***
Security and commodity brokers      0.54952   -0.2239 ***
Insurance carriers                  0.86753    0.0204 ***
Insurance brokers                   0.73771   -0.1246 ***
Bank holding companies              0.79916   -0.0384 ***
Average difference                            -0.02641

Industry                                   Regime 3

                                     Mean     Difference

Panel A. Industry Including All Postenactment Entrants

Trucking                            0.90037    0.0210 ***
Air transportation                  0.79603    0.0557 ***
Communication                       0.75682    0.0727 ***
Depository institutions             0.51167   -0.2855 ***
Nondepository credit institutions   0.63245   -0.1324 ***
Security and commodity brokers      0.43365   -0.3398 ***
Insurance carriers                  0.67494   -0.1722 ***
Insurance brokers                   0.59399   -0.2683 ***
Bank holding companies              0.59680   -0.2408 ***
Average difference                            -0.14328 **

Panel B. Industry Excluding All Postenactment Entrants

Trucking                            0.89681    0.0174 ***
Air transportation                  0.79487    0.0546 ***
Communication                       0.75420    0.0700 ***
Depository institutions             0.50209   -0.2951 ***
Nondepository credit institutions   0.62188   -0.1429 ***
Security and commodity brokers      0.42768   -0.3457 ***
Insurance carriers                  0.66557   -0.1816 ***
Insurance brokers                   0.54731   -0.3150 ***
Bank holding companies              0.63937   -0.1982 ***
Average difference                            -0.14850 **

Industry                                   Regime 4

                                     Mean     Difference

Panel A. Industry Including All Postenactment Entrants

Trucking                            0.85317   -0.0262 ***
Air transportation                  0.77025    0.0300 ***
Communication                       0.78967    0.1055 ***
Depository institutions             0.61354   -0.1836 ***
Nondepository credit institutions   0.73998   -0.0248 ***
Security and commodity brokers      0.57987   -0.1936 ***
Insurance carriers                  0.75012   -0.0970 ***
Insurance brokers                   0.69636   -0.1659 ***
Bank holding companies              0.69979   -0.1378 ***
Average difference                            -0.07706 *

Panel B. Industry Excluding All Postenactment Entrants

Trucking                            0.83045   -0.0490 ***
Air transportation                  0.76351    0.0232 **
Communication                       0.77129    0.0871 ***
Depository institutions             0.60360   -0.1936 ***
Nondepository credit institutions   0.71838   -0.0464 ***
Security and commodity brokers      0.54825   -0.2252 ***
Insurance carriers                  0.73472   -0.1124 ***
Insurance brokers                   0.70462   -0.1577 ***
Bank holding companies              0.72826   -0.1093 ***
Average difference                            -0.08702 **

*** Significant at the 0.01 level.

 ** Significant at the 0.05 level.

  * Significant at the 0.10 level.

Table VII. Descriptive Statistics on the Daily Ratio of Average
Systematic to  Average Idiosyncratic Variances per Industry

We present the mean and standard deviation for the 2,000 days
prior to the enactment date, the 2,000  days following the
enactment date, and the complete sample period of 4,000 days
centered on the relative  enactment date for the ratio of average
systematic to average idiosyncratic variances per industry
([[delta].sub.pt]):

[[delta].sub.pt] = [SYS.sub.pt]/[IDIO.sub.pt]

where [SYS.sub.pt] and [IDIO.sub.pt]  are the value-weighted
average of systematic variances and value-weighted average  of
idiosyncratic variances in industry p at time t, respectively. We
evaluate this ratio using a 300-day moving  window over the 4,000
trading days centered on the enactment date of the relevant
deregulation act for each  industry.

Industry                     Preenactment          Postenactment
                              N = 2,000             N = 2,000

                           Mean     Standard     Mean     Standard
                                    Deviation             Deviation

Panel A. Industry Including Postenactment Entrants

Trucking                  0.14393    0.09037    0.12836    0.08109
Air transportation        0.37207    0.17758    0.32082    0.09591
Communication             0.57254    0.48537    0.26885    0.08167
Depository institutions   0.28098    0.19334    0.64769    0.28891
Nondepository credit      0.32176    0.13952    0.37696    0.15655
  institutions
Security and commodity    0.30841    0.15061    0.82168    0.39725
  brokers
Insurance carriers        0.19088    0.11571    0.32341    0.18143
Insurance brokers         0.17424    0.13898    0.45714    0.25455
Bank holding companies    0.20627    0.12649    0.43072    0.19647

Panel B. Industry Excluding Postenactment Entrants

Trucking                  0.14393    0.09037    0.14779    0.10277
Air transportation        0.37207    0.17758    0.32770    0.09694
Communication             0.57254    0.48537    0.29255    0.11814
Depository institutions   0.28098    0.19334    0.67185    0.30894
Nondepository credit      0.32176    0.13952    0.40574    0.16652
  institutions
Security and commodity    0.30841    0.15061    0.88329    0.40420
  brokers
Insurance carriers        0.19088    0.11571    0.34384    0.19243
Insurance brokers         0.17424    0.13898    0.49126    0.32791
Bank holding companies    0.20627    0.12649    0.36810    0.17323

Industry                     Sample Period
                                N = 4,000

                           Mean     Standard
                                    Deviation

Panel A. Industry Including Postenactment Entrants

Trucking                  0.13615    0.08619
Air transportation        0.34644    0.14498
Communication             0.42069    0.37968
Depository institutions   0.46433    0.30666
Nondepository credit      0.34936    0.15081
  institutions
Security and commodity    0.56504    0.39510
  brokers
Insurance carriers        0.25715    0.16595
Insurance brokers         0.31569    0.24912
Bank holding companies    0.31849    0.19973

Panel B. Industry Excluding Postenactment Entrants

Trucking                  0.14586    0.09677
Air transportation        0.34988    0.14475
Communication             0.43254    0.37992
Depository institutions   0.47641    0.32342
Nondepository credit      0.36375    0.15923
  institutions
Security and commodity    0.59585    0.41911
  brokers
Insurance carriers        0.26736    0.17622
Insurance brokers         0.33275    0.29755
Bank holding companies    0.28719    0.17189

Table VIII. Changes in the Level of the Daily Ratio of Average
Systematic to Average Idiosyncratic Risk Following  Deregulation
for Each Deregulated Industry

We estimate the multivariable equation for the average ratio for
each of the nine industries:

[[delta].sub.pt] = [[alpha].sub.p] + [[gamma].sub.p](T) +
[[psi].sub.p] ([Herf.sub.p]) + [[phi].sub.1p]([D.sub.1p]) +
[[phi].sub.2p]([D.sub.2p]) + [[phi].sub.3p]([D.sub.3p]) +
[[phi].sub.4p]([D.sub.4p]) + [e.sub.pt],

where [[delta].sub.pt] is the daily ratio of average systematic
to average idiosyncratic variance per industry p, calculated on a
300-day moving window basis; thus, a higher  ratio implies higher
relative systematic variance and-or lower relative idiosyncratic
variance. We use T is a proxy for time; [Herf.sub.p] is the
quarterly Herfindahl  Index for industry p, [[psi].sub.p] is the
measure the effect of market concentration on the ratio, and
[[gamma].sub.p] measures the temporal trend in [[delta].sub.pt].
The parameters [[phi].sub.1p], [[phi].sub.2p], [[phi].sub.3p], and
[[phi].sub.4p] capture the level change in the ratio of average
systematic to average idiosyncratic variance under regimes 1, 2,
3, and 4, respectively, and [D.sub.1p], [D.sub.2p], [D.sub.3p]
and [D.sub.4p] are dummy variables for regimes 1, 2, 3, and 4,
respectively that take on the value 1 in their respective regime,
and 0 otherwise. We estimate this  equation over 4,000 trading
days centered on the enactment date of the relevant deregulation
act. We calculate the p-values based on the Newey-West standard
errors, which are in parentheses below the coefficient estimates.

Estimated                        Industry
Parameter
                 Trucking         Air        Communication
                            Transportation

[alpha]          -0.21750        0.56632        -0.12402
                 (0.000)        (0.000)         (0.089)

[gamma]           0.00009      -0.00002         -0.00058
                 (0.000)        (0.340)         (0.000)

[psi]             1.79997      -2.30315          1.56393
                 (0.000)        (0.163)         (0.067)

[[phi].sub.1p]   -0.01817       0.09670          0.34278
                 (0.320)        (0.000)         (0.000)

[[phi].sub.2p]    0.02602      -0.04929          0.56458
                 (0.197)        (0.096)         (0.000)

[[phi].sub.3p]    0.07829      -0.11877          0.80850
                 (0.000)        (0.001)         (0.000)

[[phi].sub.4p]    0.11576      -0.08062          1.16109
                 (0.000)        (0.197)         (0.000)

[R.sup.2]         0.3971         0.6336          0.1232

Estimated                          Industry
Parameter
                  Depository    Nondepository   Security and
                 Institutions       Credit        Commodity
                                 Institutions      Brokers

[alpha]            -1.13849        -0.03274        0.82971
                    (0.001)        (0.678)         (0.000)

[gamma]             0.00010        -0.00002        0.00008
                    (0.014)        (0.537)         (0.005)

[psi]              11.87289         4.96969       -4.53864
                    (0.000)        (0.000)         (0.001)

[[phi].sub.1p]     -0.39492        -0.12397       -0.06101
                    (0.000)        (0.000)         (0.183)

[[phi].sub.2p]     -0.36669        -0.19429        0.36154
                    (0.000)        (0.000)         (0.000)

[[phi].sub.3p]      0.14514         0.16232        0.86218
                    (0.031)        (0.000)         (0.000)

[[phi].sub.4p]     -0.14841        -0.13596        0.24671
                    (0.044)        (0.008)         (0.018)

[R.sup.2]            0.6474         0.4241          0.6418

Estimated                     Industry
Parameter
                 Insurance   Insurance      Bank
                  Carriers    Brokers     Holding
                                         Companies

[alpha]            0.24707     0.20869     0.39163
                  (0.000)     (0.000)     (0.000)

[gamma]            0.00008     0.00009     0.00012
                  (0.000)     (0.000)     (0.000)

[psi]              0.79344     0.38007    -1.00374
                  (0.031)     (0.022)     (0.269)

[[phi].sub.1p]    -0.07329    -0.03449    -0.14776
                  (0.002)     (0.327)     (0.000)

[[phi].sub.2p]    -0.16766    -0.01186    -0.05127
                  (0.000)     (0.758)     (0.234)

[[phi].sub.3p]     0.17014     0.36290     0.26137
                  (0.001)     (0.000)     (0.000)

[[phi].sub.4p]    -0.04697     0.06460    -0.07348
                  (0.396)     (0.423)     (0.350)

[R.sup.2]          0.4265      0.4981      0.6595

Table IX. Changes in the Level of the Daily Ratio of Average
Systematic to Average Idiosyncratic Risk Following  Deregulation
for Each Deregulated Industry, Excluding Postenactment New
Entrants

We estimate the multivariable equation for the average ratio for
each of the nine industries in our study after excluding
postenactment new entrants:

[[delta].sub.pt] = [[alpha].sub.p] + [[gamma].sub.p](T) +
[[psi].sub.p] ([Herf.sub.p]) + [[phi].sub.1p]([D.sub.1p]) +
[[phi].sub.2p]([D.sub.2p]) + [[phi].sub.3p]([D.sub.3p]) +
[[phi].sub.4p]([D.sub.4p]) + [e.sub.pt],

where [[delta].sub.p] is the daily ratio of average systematic to
average idiosyncratic variance per industry p (excluding
postenactment new entrants), which we calculate  on a 300-day
moving window basis; thus, a higher ratio implies higher relative
systematic variance and-or lower relative idiosyncratic variance.
We use T as  a proxy for time and [Herf.sub.p] is the quarterly
Herfindahl Index for industry p. Therefore, [[psi].sub.p]
measures the effect of market concentration on the ratio, and
[[gamma].sub.p]  measures the temporal trend in [[delta].sub.pt].
The parameters [[phi].sub.1p], [[phi].sub.2p], [[phi].sub.3p],
and [[phi].sub.4p] capture the level change in the ratio of
average systematic to average idiosyncratic  variance under
regimes 1, 2, 3, and 4, respectively, and, [D.sub.1p],
[D.sub.2p], [D.sub.3p] and [D.sub.4p] are dummy variables for
regimes 1, 2, 3, and 4, respectively, that take on the value 1
in their respective regime, and 0 otherwise. We estimate this
equation over 4,000 trading days centered on the enactment date
of the relevant deregulation act.  We calculate the p-values
based on the Newey-West standard errors, which we report in
parentheses below the coefficient estimates.

Estimated                        Industry
Parameter
                 Trucking         Air        Communication
                            Transportation

[alpha]          -0.23815       0.57754         -0.11615
                  (0.000)       (0.000)         (0.113)

[gamma]           0.00011      -0.00002         -0.00057
                  (0.000)       (0.323)         (0.000)

[psi]             1.94516      -2.43108          1.63509
                  (0.000)       (0.148)         (0.058)

[[phi].sub.1p]   -0.02208       0.09669          0.33142
                  (0.249)       (0.000)         (0.000)

[[phi].sub.2p]    0.02209      -0.04886          0.55673
                  (0.284)       (0.099)         (0.000)

[[phi].sub.3p]    0.07682      -0.11781          0.79205
                  (0.000)       (0.001)         (0.000)

[[phi].sub.4p]    0.13619      -0.07043          1.16926
                  (0.000)       (0.149)         (0.000)

[R.sup.2]          0.4521        0.5909          0.1111

Estimated                          Industry
Parameter
                  Depository    Nondepository   Security and
                 Institutions       Credit        Commodity
                                 Institutions      Brokers

[alpha]            -1.34109        -0.01232        0.66313
                    (0.000)        (0.888)         (0.000)

[gamma]             0.00007        -0.00001        0.00010
                    (0.086)        (0.838)         (0.000)

[psi]              13.26281         4.82711       -2.59088
                    (0.000)        (0.000)         (0.060)

[[phi].sub.1p]     -0.41163        -0.12871       -0.04963
                    (0.000)        (0.000)         (0.269)

[[phi].sub.2p]     -0.37375        -0.19973        0.35540
                    (0.000)        (0.000)         (0.000)

[[phi].sub.3p]      0.18527         0.17633        0.87047
                    (0.011)        (0.000)         (0.000)

[[phi].sub.4p]     -0.09156        -0.11398        0.30553
                    (0.239)        (0.032)         (0.003)

[R.sup.2]            0.6474         0.4546          0.6774

Estimated                     Industry
Parameter
                 Insurance   Insurance      Bank
                  Carriers    Brokers     Holding
                                         Companies

[alpha]            0.24704     0.19929     0.38837
                  (0.000)     (0.000)     (0.000)

[gamma]            0.00008     0.00007     0.00013
                  (0.000)     (0.011)     (0.000)

[psi]              0.87952     0.28169    -0.84921
                  (0.239)     (0.072)     (0.315)

[[phi].sub.1p]    -0.07589     0.00217    -0.16487
                   (0.001     (0.958)     (0.000)

[[phi].sub.2p]    -0.16734     0.07848    -0.14567
                  (0.000)     (0.092)     (0.000)

[[phi].sub.3p]     0.18982     0.56937     0.14309
                   (0.000     (0.000)     (0.008)

[[phi].sub.4p]    -0.02566     0.13148    -0.14507
                  (0.651)     (0.186)     (0.047)

[R.sup.2]          0.4621      0.4694      0.5972

Table X. Pre-and Postderegulation Means of Industry-Specific
Variance Divided  by Total Industry Variance

Table values are the mean values of the industry-specific
variance, divided by total industry variance for  the pre-and
postderegulation periods. We use Newey-West standard errors to
determine the significance of  the difference in the means for
each industry and then present the average difference across
industries. We  estimate this equation over 600 days (Panel A)
and 4,000 days (Panel B) centered on the enactment date of  the
relevant deregulation act.

Industry                             Preenactment      Postenactment

                                      Mean      N      Mean       N

      Panel A. Results Estimated Over 600 Days Centered on the
                            Enactment Date

Trucking                            0.71728    300    0.82971    300
Air transportation                  0.71087    300    0.53491    300
Communication                       0.51220    300    0.53144    300
Depository institutions             0.24325    300    0.47051    300
Nondepository credit institutions   0.31685    300    0.55415    300
Security and commodity brokers      0.36220    300    0.50110    300
Insurance carriers                  0.28084    300    0.57592    300
Insurance brokers                   0.47983    300    0.70109    300
Bank holding companies              0.22749    300    0.28396    300
Average difference

     Panel B. Results Estimated Over 4, 000 Days Centered on
                            the Enactment Date

Trucking                            0.70520   2,000   0.75181   2,000
Air transportation                  0.65164   2,000   0.61723   2,000
Communication                       0.36887   2,000   0.47288   2,000
Depository institutions             0.38291   2,000   0.26929   2,000
Nondepository credit institutions   0.43719   2,000   0.46930   2,000
Security and commodity brokers      0.45565   2,000   0.31551   2,000
Insurance carriers                  0.37486   2,000   0.41039   2,000
Insurance brokers                   0.67104   2,000   0.50780   2,000
Bank holding companies              0.28434   2,000   0.22612   2,000
Average difference

Industry                             Difference

   Panel A. Results Estimated Over 600 Days
        Centered on the Enactment Date

Trucking                             0.11243 ***
Air transportation                  -0.17596 ***
Communication                        0.01925 **
Depository institutions              0.22726 ***
Nondepository credit institutions    0.23730 ***
Security and commodity brokers       0.13890 ***
Insurance carriers                   0.29508 ***
Insurance brokers                    0.22126 ***
Bank holding companies               0.05647 ***
Average difference                   0.12578 **

    Panel B. Results Estimated Over 4, 000
      Days Centered on the Enactment Date

Trucking                             0.04661 ***
Air transportation                  -0.03441 ***
Communication                        0.10400 ***
Depository institutions             -0.11362 ***
Nondepository credit institutions    0.03212 **
Security and commodity brokers      -0.14014 ***
Insurance carriers                   0.03554 **
Insurance brokers                   -0.16324 ***
Bank holding companies              -0.05822 ***
Average difference                     -0.03237

*** Significant at the 0.01 level.

 ** Significant at the 0.05 level.

Table XI. Pre- and Postderegulation Means of Industry-Specific
Variance Divided  by Total Industry Variance Excluding
Postenactment Entrants

Table values are the mean values of the industry-specific
variance, divided by total industry variance for  the pre- and
postderegulation periods. We use Newey-West standard errors to
determine the significance of  the difference in the means for
each industry and then present the average difference across
industries. We  estimate this equation over 600 days (Panel A)
and 4,000 days (Panel B) centered on the enactment date of  the
relevant deregulation act.

Industry                            Preenactment      Postenactment

                                     Mean       N      Mean       N

       Panel A. Results Estimated Over 600 Days Centered on the
                          Enactment Date

Trucking                            0.71728    300    0.82726    300
Air transportation                  0.71087    300    0.53491    300
Communication                       0.51220    300    0.53181    300
Depository institutions             0.24325    300    0.46856    300
Nondepository credit institutions   0.31685    300    0.54798    300
Security and commodity brokers      0.36220    300    0.50111    300
Insurance carriers                  0.28084    300    0.57259    300
Insurance brokers                   0.47983    300    0.68999    300
Bank holding companies              0.22749    300    0.29271    300
Average difference

     Panel B. Results Estimated Over 4,000 Days Centered on the
                          Enactment Date

Trucking                            0.70520   2,000   0.74337   2,000
Air transportation                  0.65164   2,000   0.62088   2,000
Communication                       0.36887   2,000   0.52833   2,000
Depository institutions             0.38291   2,000   0.28542   2,000
Nondepository credit institutions   0.43719   2,000   0.45810   2,000
Security and commodity brokers      0.45565   2,000   0.29520   2,000
Insurance carriers                  0.37486   2,000   0.41073   2,000
Insurance brokers                   0.67104   2,000   0.49539   2,000
Bank holding companies              0.28434   2,000   0.25636   2,000
Average difference

Industry                            Difference

    Panel A. Results Estimated Over 600 Days
       Centered on the Enactment Date

Trucking                             0.10999 **
Air transportation                  -0.17596 ***
Communication                        0.01961 ***
Depository institutions              0.22531 ***
Nondepository credit institutions    0.23113 ***
Security and commodity brokers       0.13891 ***
Insurance carriers                   0.29175 ***
Insurance brokers                    0.21016 ***
Bank holding companies               0.06523 ***
Average difference                   0.12401 **

   Panel B. Results Estimated Over 4,000 Days
      Centered on the Enactment Date

Trucking                             0.03818 ***
Air transportation                  -0.03076 ***
Communication                        0.15946 ***
Depository institutions             -0.09749 ***
Nondepository credit institutions    0.02091 *
Security and commodity brokers      -0.16045 ***
Insurance carriers                   0.03587 **
Insurance brokers                   -0.17565 ***
Bank holding companies              -0.02798 ***
Average difference                  -0.02644

*** Significant at the 0.01 level.

 ** Significant at the 0.05 level.

  * Significant at the 0.10 level.

Table XII. Means of Industry-Specific Variance Divided by Total
Industry Variance Prior to the Deregulation and Under  the Four
Regimes Following Deregulation

Table values are the mean value of the industry-specific variance
divided by total industry variance for the prederegulation period
and across each of the  four regimes following deregulation. The
differences in this table are the differences between the
respective regime and the prederegulation period. We use
Newey-West standard errors to determine the significance of the
difference in the means between the respective regime and the
prederegulation period. We  present the average difference across
industries for each regime. We estimate the mean for each
industry portfolio (Panel A) and for each industry portfolio
excluding postenactment entrants (Panel B)

Industry                            Preenactment

                                        Mean

Panel A. Industry Portfolio Including All Postenactment Entrants

Trucking                              0.70520
Air transportation                    0.65164
Communication                         0.36887
Depository institutions               0.38291
Nondepository credit institutions     0.43719
Security and commodity brokers        0.45565
Insurance carriers                    0.37486
Insurance brokers                     0.67104
Bank holding companies                0.28434
Average difference

Panel B. Industry Portfolio Excluding All Postenactment Entrants

Trucking                              0.70520
Air transportation                    0.65164
Communication                         0.36887
Depository institutions               0.38291
Nondepository credit institutions     0.43719
Security and commodity brokers        0.45565
Insurance carriers                    0.37486
Insurance brokers                     0.67104
Bank holding companies                0.28434
Average difference

Industry                                   Regime 1

                                     Mean     Difference

Panel A. Industry Portfolio Including All Postenactment Entrants

Trucking                            0.82963    0.1244 ***
Air transportation                  0.53500   -0.1166 ***
Communication                       0.53142    0.1625 ***
Depository institutions             0.47029    0.0874 ***
Nondepository credit institutions   0.55375    0.1166 ***
Security and commodity brokers      0.50115    0.0455 ***
Insurance carriers                  0.57550    0.2006 ***
Insurance brokers                   0.70095    0.0299 **
Bank holding companies              0.28391   -0.0004
Average difference                             0.07221 **

Panel B. Industry Portfolio Excluding All Postenactment Entrants

Trucking                            0.81908    0.1139 ***
Air transportation                  0.53500   -0.1166 ***
Communication                       0.53191    0.1630 ***
Depository institutions             0.47153    0.0886 ***
Nondepository credit institutions   0.52735    0.0902 ***
Security and commodity brokers      0.47706    0.0214 **
Insurance carriers                  0.55872    0.1839 ***
Insurance brokers                   0.65446   -0.0166 ***
Bank holding companies              0.29294    0.00859
Average difference                             0.05959 *

Industry                                   Regime 2

                                     Mean     Difference

Panel A. Industry Portfolio Including All Postenactment Entrants

Trucking                            0.87075    0.1656 ***
Air transportation                  0.63252   -0.0191 **
Communication                       0.48232    0.1134 ***
Depository institutions             0.41046    0.0276 *
Nondepository credit institutions   0.68728    0.2501 ***
Security and commodity brokers      0.35405   -0.1016 ***
Insurance carriers                  0.67251    0.2977 ***
Insurance brokers                   0.63266   -0.0384 **
Bank holding companies              0.26644   -0.0179 **
Average difference                             0.07525

Panel B. Industry Portfolio Excluding All Postenactment Entrants

Trucking                            0.86045    0.1553 ***
Air transportation                  0.63296   -0.0187 **
Communication                       0.48860    0.1197 ***
Depository institutions             0.42054    0.0376 **
Nondepository credit institutions   0.66143    0.2242 ***
Security and commodity brokers      0.33391   -0.1217 ***
Insurance carriers                  0.65690    0.2820 ***
Insurance brokers                   0.57502   -0.0960 ***
Bank holding companies              0.28451    0.0002 *
Average difference                             0.06474

Industry                                   Regime 3

                                     Mean     Difference

Panel A. Industry Portfolio Including All Postenactment Entrants

Trucking                            0.75619    0.0510 ***
Air transportation                  0.66346    0.0118 *
Communication                       0.42121    0.0523 ***
Depository institutions             0.18610   -0.1968 ***
Nondepository credit institutions   0.40519   -0.0320 *
Security and commodity brokers      0.23421   -0.2214 ***
Insurance carriers                  0.35154   -0.0233 ***
Insurance brokers                   0.36371   -0.3073 ***
Bank holding companies              0.14035   -0.1440 ***
Average difference                            -0.08997 *

Panel B. Industry Portfolio Excluding All Postenactment Entrants

Trucking                            0.74488    0.0397 ***
Air transportation                  0.66886    0.0172 **
Communication                       0.47976    0.1109 ***
Depository institutions             0.18891   -0.1940 ***
Nondepository credit institutions   0.38472   -0.0525 **
Security and commodity brokers      0.21293   -0.2427 ***
Insurance carriers                  0.34743   -0.0274 **
Insurance brokers                   0.31115   -0.3599 ***
Bank holding companies              0.15993   -0.1244 ***
Average difference                            -0.09257 *

Industry                                   Regime 4

                                     Mean     Difference

Panel A. Industry Portfolio Including All Postenactment Entrants

Trucking                            0.69708   -0.0081
Air transportation                  0.62280   -0.0288 ***
Communication                       0.46848    0.0996 ***
Depository institutions             0.19891   -0.1840 ***
Nondepository credit institutions   0.40445   -0.0327 **
Security and commodity brokers      0.27674   -0.1789 ***
Insurance carriers                  0.31017   -0.0647 ***
Insurance brokers                   0.46059   -0.2105 ***
Bank holding companies              0.22280   -0.0615 ***
Average difference                            -0.07441 **

Panel B. Industry Portfolio Excluding All Postenactment Entrants

Trucking                            0.69051   -0.0147 *
Air transportation                  0.62783   -0.0238 **
Communication                       0.55141    0.1825 ***
Depository institutions             0.22436   -0.1586 ***
Nondepository credit institutions   0.40388   -0.0333 ***
Security and commodity brokers      0.25767   -0.1980 ***
Insurance carriers                  0.32071   -0.0542 ***
Insurance brokers                   0.48070   -0.1903 ***
Bank holding companies              0.26503   -0.0193 *
Average difference                            -0.05662

*** Significant at the 0.01 level.

 ** Significant at the 0.05 level.

  * Significant at the 0.10 level.

Table XIII. Changes in the Level of the Daily Ratio of Industry
Systematic to Industry-Specific Risk Following  Deregulation

We estimate the multivariable equation for the nine
value-weighted industry portfolios in our study:

[[delta].sub.It] = [a.sub.I] + [gamma].sub.I](T) +
[[psi].sub.I]([Herf.sub.I]) + [[phi].sub.1I]([D.sub.1I]) +
[[phi].sub.2I]([D.sub.2I]) + [[phi].sub.3I]([D.sub.3I]) +
[[phi].sub.4I]([D.sub.4I]) + [e.sub.It],

where [[summation].sub.It] is the daily ratio of industry
systematic to industry-specific variance for industry portfolio
I, calculated on a 300-day moving window basis; thus,  a higher
ratio implies higher relative systematic variance and-or lower
relative idiosyncratic variance. We use T as a proxy for time and
[Herf.sub.I] as the quarterly  Herfindahl Index for portfolio i.
Therefore, [[psi].sub.I] captures the effect of market
concentration on the ratio, and y, measures the temporal trend in
[[delta].sub.It]. The  parameters [[phi].sub.1I], [[phi].sub.2I],
[[phi].sub.3I], and [[phi].sub.4I], capture the
level change in the ratio of systematic to idiosyncratic variance
under regimes 1, 2, 3, and 4, respectively, and  [D.sub.1I],
[D.sub.2I], [D.sub.13], and [D.sub.4I], are dummy variables for
regimes 1, 2, 3, and 4, respectively, that take on the value 1 in
their respective regime, and 0 otherwise. We  estimate this
equation over 4,000 trading days centered on the enactment date
of the relevant deregulation act, and calculate the p-values based
on the Newey-West  standard errors, which are in parentheses
below the coefficient estimates.

Estimated                      Portfolio
Parameter
                Trucking         Air        Communication
                           Transportation

[alpha]         -0.59592       0.72273          3.70908
                 (0.000)       (0.007)         (0.000)

[gamma]          0.00014       0.00001         -0.00126
                 (0.000)       (0.743)         (0.000)

[psi]            3.56800      -1.72345          4.51161
                 (0.000)       (0.505)         (0.056)

[[phi].sub.1]   -0.05996       0.27737          0.28288
                 (0.118)       (0.000)         (0.012)

[[phi].sub.2]   -0.00208      -0.01235          0.90833
                 (0.964)       (0.775)         (0.000)

[[phi].sub.3]    0.15443      -0.11317          1.56808
                 (0.002)       (0.027)         (0.000)

[[phi].sub.4]    0.27236      -0.02299          2.32808
                 (0.000)       (0.763)         (0.000)
[R.sup.2]]        0.3864        0.1406          0.5614

Estimated                           Portfolio
Parameter
                 Depository    Nondepository   Security and
                Institutions       Credit        Commodity
                                Institutions      Brokers

[alpha]           -7.70322         0.01675       -0.03416
                   (0.000)        (0.915)         (0.887)

[gamma]            0.00006         0.00002        0.00030
                   (0.673)        (0.820)         (0.000)

[psi]             74.26839        20.46566        9.86997
                   (0.000)         (0.000         (0.000)

[[phi].sub.1]     -3.27406        -1.29481       -0.42517
                   (0.000)        (0.000)         (0.000)

[[phi].sub.2]     -3.09811        -1.60258        0.29474
                   (0.000)        (0.000)         (0.048)

[[phi].sub.3]      0.28811        -0.44844        1.72207
                   (0.419)        (0.006)         (0.000)

[[phi].sub.4]      0.45342        -0.77409        0.99878
                   (0.141)        (0.000)         (0.001)
[R.sup.2]]          0.7000         0.5048          0.5816

Estimated                   Portfolio
Parameter
                Insurance   Insurance      Bank
                 Carriers    Brokers     Holding
                                        Companies

[alpha]           1.39271     0.20223     2.29793
                 (0.000)     (0.033)     (0.014)

[gamma]           0.00020     0.00016     0.00028
                 (0.103)     (0.002)     (0.391)

[psi]            13.43350     1.32120     3.95298
                 (0.005)     (0.013)     (0.649)

[[phi].sub.1]    -1.59156    -0.46058    -0.62233
                 (0.000)     (0.000)     (0.032)

[[phi].sub.2]    -1.98621    -0.26818    -0.42315
                 (0.000)     (0.006)     (0.318)

[[phi].sub.3]    -0.39284     0.94003     3.10677
                 (0.178)     (0.000)     (0.000)

[[phi].sub.4]    -0.17971     0.44139     0.88122
                 (0.592)     (0.024)     (0.323)
[R.sup.2]]        0.3233      0.4698      0.3514

Table XIV. Changes in the Level of the Daily Ratio of Industry
Systematic to Industry-Specific Risk Following  Deregulation for
Industry Portfolios Excluding Postenactment Entrants

We estimate the multivariable equation for the nine
value-weighted industry portfolios comprised only of firms that
existed prior to the enactment date:

[[delta].sub.It] = [a.sub.I] + [gamma].sub.I](T) +
[[psi].sub.I]([Herf.sub.I]) + [[phi].sub.1I]([D.sub.1I]) +
[[phi].sub.2I]([D.sub.2I]) +  [[phi].sub.3I]([D.sub.3I]) +
[[phi].sub.4I]([D.sub.4I]) + [e.sub.It],

where [[delta].sub.It] is the daily ratio of industry systematic
to industry-specific variance for industry portfolio I,
calculated on a 300-day moving window basis; thus, a  higher
ratio implies higher relative systematic variance and-or lower
relative idiosyncratic variance. We use T as a proxy for time,
and [Herf.sub.I] is the quarterly  Herfindahl Index for portfolio
i, [[psi].sub.I] to measure the effect of market concentration on
the ratio, and [[gamma].sub.I] measures the temporal trend in
[[delta].sub.It]. The parameters [[phi].sub.1I], [[phi].sub.2I],
[[phi].sub.2I], and [[phi].sub.4I], capture the level change in
the ratio of systematic to idiosyncratic variance under regimes
1, 2, 3, and 4, respectively, and  [D.sub.1I], [D.sub.2I],
[D.sub.3I], and [D.sub.4I] are dummy variables for regimes 1,
2, 3, and 4, respectively, that take on the value 1 in their
respective regime, and 0 otherwise. We estimate this equation
over  4,000 trading days centered on the enactment date of the
relevant deregulation act and calculate the p-values based on the
Newey-West standard errors, which  are in parentheses below the
coefficient estimates.

Estimated                      Portfolio
Parameter
                Trucking         Air        Communication
                           Transportation

[alpha]         -0.67808       0.77712          3.68175
                 (0.000)       (0.003)         (0.000)

[gamma]          0.00017       0.00001         -0.00125
                 (0.000)       (0.758)         (0.000)

[psi]            3.64140      -2.29560          4.42910
                 (0.000)       (0.368)         (0.056)

[[phi].sub.1]   -0.06370       0.27407          0.28639
                 (0.098)       (0.000)         (0.010)

[[phi].sub.2]   -0.01193      -0.01553          0.87925
                 (0.793)       (0.720)         (0.000)

[[phi].sub.3]    0.13604      -0.13456          1.27528
                 (0.006)       (0.008)         (0.000)

[[phi].sub.4]    0.22063      -0.05190          2.02273

                 (0.000)       (0.487)         (0.000)

[R.sup.2]         0.3937        0.1449          0.6020

Estimated                         Portfolio
Parameter
                 Depository    Nondepository   Security and
                Institutions       Credit        Commodity
                                Institutions      Brokers

[alpha]           -9.02313        -0.09910        0.20900
                   (0.000)        (0.463)         (0.367)

[gamma]           -0.00014        -0.00003        0.00030
                   (0.311)        (0.743)         (0.000)

[psi]             86.31805        21.13800        6.38244
                   (0.000)        (0.000)         (0.000)

[[phi].sub.1]     -3.43614        -1.11866       -0.39627
                   (0.000)        (0.000)         (0.000)

[[phi].sub.2]     -3.28847        -1.40893        0.31858
                   (0.000)        (0.000)         (0.034)

[[phi].sub.3]      0.17696        -0.28748        1.73703
                   (0.625)        (0.057)         (0.000)

[[phi].sub.4]      0.16866        -0.71284        0.96914
                   (0.597)        (0.000)         (0.001)

[R.sup.2]           0.6339         0.4992          0.5850

Estimated                   Portfolio
Parameter
                Insurance   Insurance      Bank
                 Carriers    Brokers     Holding
                                        Companies

[alpha]           1.26093     0.12260     3.13213
                 (0.000)     (0.210)     (0.000)

[gamma]           0.00020     0.00018    -0.00010
                 (0.115)     (0.001)     (0.732)

[psi]            14.81894     1.19766    -0.68165
                 (0.001)     (0.020)     (0.933)

[[phi].sub.1]    -1.49191    -0.37544    -0.35409
                 (0.000)     (0.000)     (0.208)

[[phi].sub.2]    -1.89600    -0.16169    -0.16153
                 (0.000)     (0.116)     (0.689)

[[phi].sub.3]    -0.41567     1.13328     2.96646
                 (0.140)     (0.000)     (0.000)

[[phi].sub.4]    -0.34802     0.21413     0.83521
                 (0.294)     (0.316)     (0.320)

[R.sup.2]         0.2861      0.4416      0.2368

Figure 2. Moving Ratio of Average Systematic to Average
Idiosyncratic  Variances Centered on the Relevant Enactment Date

We estimate daily average ratios using a 300-day moving window on
each portfolio. The zero value on the  x-axis (relative time)
represents the enactment date.

Relative to deregulation event

Regime   First trading   Last trading
              day            day

1              1             300
2             301            600
3             601            900
4             901           2,000
COPYRIGHT 2011 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Semaan, Elias; Drake, Pamela Peterson
Publication:Financial Management
Geographic Code:1USA
Date:Jun 22, 2011
Words:16384
Previous Article:The performance of private equity-backed IPOs.
Next Article:Resolving the presidential puzzle.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters