 |
 |
|
|
| Back to Current Developments |
|
|
EMPLOYMENT ARBITRATION
WHAT DOES THE DATA SHOW? Arbitration of statutory employment disputes is highly
controversial. Everyone seems
to have an opinion on the subject.
The National Employment Lawyers’ Association compares employment
arbitration to the kangaroo courts of the Soviet Union.
The Equal Employment Advisory Council (an employer group) calls
employment arbitration “fair, efficient, and less costly”. What is largely missing from this debate is any reference
to the facts. Despite the
publication of several scholarly articles analyzing the outcomes of
employment arbitration, most advocates continue to rely on assumptions. The purpose of this presentation is not to add yet
another opinion to the debate. Rather,
it is to summarize the currently available data to provide a more
informed basis for policy making. I. EMPLOYEE WIN RATE
The most important question regarding employment
arbitration is whether employees with legitimate claims are able to
prevail. Critics of arbitration assert that it is biased
against employees. These critics
support their claim by pointing out that many employer plans do not
contain provisions providing for important due process protections. They are also concerned that employment arbitration
systems are designed by one of the parties to the dispute, a blatant
conflict of interest. Both these concerns, and many others, are legitimate. But they are concerns, not facts. Ultimately, employees either win their cases in arbitration or they do not. The only way to know is to look at the results. There have been several empirical studies of the
rate at which employees prevail in arbitration.
Each of these studies was well conducted, and there
is no reason to consider any of them to be more valid than the others. Combining the data sets of all four articles
into a single data set and computing the success rate for the entire
group yields a success rate of 62%. Standing alone, this result has little meaning. The critical question is, “how does the employee
success rate in arbitration compare with the employee success rate
in court?”.
There have been several studies of employees’ success
rate in court against their employers.
All but one, however, used what has recently been discovered
to be an incorrect methodology. In
any comparison, it is essential to “compare apples to apples”.
The arbitration data sets that have been studied
consist primarily of contract cases.
Therefore, the results must be compared to the results of contract
cases brought by employees. This
is especially important in light of the finding by Eisenberg and Hill
that employees win civil rights cases far less often than contract
cases.1 The only study to date using the correct methodology
was conducted by Eisenberg and Hill.2 They found that employees prevailed in 57%
of their cases. While this
is slightly lower than the 63% success rate in arbitration, the difference
is not statistically significant. Not all employment cases, however, result in a trial. Many are dismissed upon summary judgment.
The majority of civil rights claims, 60%, are dismissed on
summary judgment.3 For contract claims, the rate of summary judgment
is 15%.4 The rate of summary judgment for the mix of cases involved in the
relevant studies is 25%. This
means that 25% of the cases of the cases that went to arbitration
would have been dismissed had they gone to court. In arbitration, by contrast, none of the cases studied were resolved by summary judgment.5 When you look at the results of all cases employees
bring (including the ones dismissed on summary judgment), the rate
of success is 43%. This is
dramatically lower than the 63% success rate of employees in arbitration.
It is important to be absolutely clear on what these
numbers do and do not mean. All
the studies of arbitration results have been conducted on data from
the American Arbitration Association (AAA).
AAA is the only arbitration provider whose caseload is both
large enough to study and organized in a manner that researchers can
use. The greater success rate of employees in arbitration
means, therefore, that employees win more often in a fair arbitration. It does not show that this is generally the
case in practice. AAA is the
oldest and largest arbitration provider.
Its arbitrators are rigorously screened for conflict of interest. All AAA cases are conducted in compliance with
the ABA Protocol. These conditions may also exist with other major
providers, such as the National Academy of Arbitrators, National Arbitration
Forum, and JAMS/Endispute. The
results of arbitrations conducted by these groups, however, have not
been analyzed. Even more important, there are literally hundreds
of other arbitration providers. The
field is entirely unregulated. It
is completely unwarranted to assume that the results of all employment
arbitration are comparable to those of AAA. II. SIZE OF JUDGMENTS
It is not sufficient to know how often employees
prevail in arbitration. The
size of the judgment is equally important.
An employee who has a $100,000 claim but receives only $10,000,
has not received justice. The data on the size of arbitration awards is very
intriguing. Eisenberg and Hill found that the median award
for employees who prevail in arbitration is $63,120. The median award for those who prevail in court
is $68,737. This difference
is not statistically significant. AWARDS
When they examined the mean awards, however, a different
picture emerged. The mean
award in arbitration is $153,000.
The mean jury award, however, is $462,000 (over three times
as much). These figures, however, do not represent the amount
actually received by successful employees.
In litigation, an employer who loses has the right to appeal. The risk of losing the appeal, plus the additional
time and expense, leads many employees who have prevailed at trial
to settle their case for much less than the jury awarded. The best available estimate is that amount
ultimately received by successful employees is about half of the jury
award6, or $231,000. Because it is extremely difficult to appeal an arbitration award,
this dynamic is largely absent in arbitration.
Even with this refinement, however, mean litigation awards
are much higher.
This data suggests that the conventional wisdom about
arbitrators and damages may be correct.
Arbitrators appear to award comparable damages in most cases,
but to award less in large cases, where the majority of the damages
are non-economic. The most recent analysis, however, casts doubt upon
this conclusion. Damages awarded
by arbitrators and courts should only be equal if the damages suffered
by the employees are equal. In
the absence of any reason to believe they are different, researchers
have always assumed comparable actual damages. The Institute’s 2003 study, however, indicated that
the financial hurdles to bringing a case are lower in arbitration
than in court.7 (This
will be discussed in more detail in Section III.) This results in cases being brought in arbitration
with damages lower than the cases that go to court. Dropping these cases from the arbitration data
set and examining the results of arbitration cases where the damages
were large enough to justify litigation produces mean damages in arbitration
of $292,0008 (slightly larger than mean damages in litigation).
In light of the fact that the 50% reduction from
jury award to actual award is only an informed estimate, one cannot
infer that mean arbitration awards in comparable cases are actually
higher, only that previous analyses that suggested that juries were
more generous in large cases may be incorrect.
The bottom line regarding the relative size of jury
awards versus arbitration awards today is that there is no bottom
line. The experience of trial lawyers (on both sides)
suggests that juries may be more easily convinced to award large non-economic
damages. There is some data
to support this assumption. But
other data suggests that this is not the case. More research is needed before we will know the answer. III. ACCESS TO JUSTICE
Supporters of employment arbitration claim that it
increases the number of aggrieved employees who are able to obtain
justice. They argue that many worthy employment claims
are never brought because the employee cannot afford to litigate and
the potential recovery is too small for an attorney to accept on a
contingency agreement. Arbitration,
because it is less expensive, provides access to justice for many
who otherwise get none. This is a plausible theory, but is it sustained by
the data? The Institute conducted an initial analysis of this
question. The minimum damages
required to sustain employment litigation is $75,000.9 Many would argue that the number is much higher.
If employees are able to bring arbitration claims where the
damages are less than $75,000, it would support the theory that arbitration
increases access to justice. The Institute looked at all AAA employment arbitrations
for the year 2000 for which there was a stated demand. In the majority of these cases (54%), the demand was less than $75,000.
Many cases (26%) involved claims of less than $25,000.10 In other words, half of the people whose employment
claims were heard by AAA that year would not have been able to bring
their claims to court. The same indication emerges from Hill’s analysis
of 200 AAA employment cases. She
found that the median demand where the employee was represented by
counsel was $75,000.11 This
is the minimum demand for a case to be litigated. Again, half of AAA cases involved demands too
small to litigate. These studies, while far from conclusive, both indicate
that arbitration really does increase access to justice for employees
with smaller claims. IV. CONCLUSION
We don’t have final answers to the critical questions
about employment arbitration. The
number of studies, and the size of the data sets involved, is too
small to draw conclusions. But there is a significant amount of data, which
has been analyzed by several highly qualified researchers. This data consistently points in the same direction. Research to date indicates that more employees
are able to gain access to justice through arbitration than through
litigation, and that they are more likely to win their cases in arbitration
(if they use a qualified arbitration provider). The picture on damages is more cloudy. The data indicates that most employees receive
awards in arbitration comparable to what they would have received
in court. Whether or not employees
with large claims fare as well in arbitration as in court remains
to be seen. Clearly, continued research is needed on this vitally
important issue. The National
Workrights Institute intends to be involved in this research and to
share the results with policy makers. 1 Theodore Eisenberg & Elizabeth Hill, Employment Arbitration and Litigation: An Empirical Comparison, 2003 Pub. L. & Legal Theory Res. Paper Series 1, 14, available at http://papers.ssrn.com/sol3/papers.crm?abstract_id=389780 (last visited Oct. 6, 2003). 2 See id. 3 Search of Inter-University Consortium for Political and Social Research Database, case category 442 jobs (July 11, 1997) (on file with author). 4 78% of AAA’s employment docket consists of cases that are dismissed on summary judgment by courts at a rate of 15%. See id. This means that 11.7% of AAA’s docket would have been dismissed on summary judgment if the cases had gone to court. The remaining 22% of AAA’s employment docket consists of cases that are dismissed on summary judgment, at a rate of 60%, by courts. See id. This means that an additional 13.2% of AAA’s cases would have been dismissed on summary judgment had they gone to court. Thus, a total of 24.9% of AAA’s employment would not have survived summary judgment. 5 Lewis L. Maltby, The Myth of Second-Class Justice: Resolving Employment Disputes in Arbitration, in How ADR Works 915, 921 (Norman Brand ed., 2002). 6 Lewis L. Maltby, Employment Arbitration and Workplace Justice, 38 U.S.F. L. Rev. 105 (2003). 7 Lewis L. Maltby, Out of the Frying Pan, Into the Fire: The Feasibility of Post-Dispute Employment Arbitration Agreements, 30 Wm. Mitchell L. Rev. 313 (2003). 8 These calculations were conducted as part of this report. 9 William M. Howard, Arbitrating Claims of Employment Discrimination, Dis. Resol. J., Oct.-Dec. 1995, at 40. See also William M. Howard, Mandatory Arbitration of Employment Discrimination Disputes (1995) (unpublished Ph.D. dissertation, Arizona State University). 10 Lewis L. Maltby, Arbitrating Employment Disputes: The Promise and the Peril, in Arbitration of Employment Disputes 530 (Danial P. O’Meara ed., 2002). 11 Elizabeth Hill, Due Process at Low Cost: An Empirical Study of employment Arbitration Under the Auspices of the American Arbitration Association, 18 Ohio St. J. on Disp. Resol. 777 (2003). Back to top |
|
| Back to Current Developments |
|
|
|
The National Workrights Institute
166 Wall Street, Princeton, NJ 08540
(609) 683 0313
info@workrights.org
|