Applying Econometrics to Estimate Damages

• Pages: 301-340

301

CHAPTER 12

APPLYING ECONOMETRICS TO ESTIMATE

DAMAGES

The calculation of damages requires estimating the economic impact

of an allegedly anticompetitive act. A damages analysis, however, must

isolate the impact of an allegedly illegal act from other potentially causal

factors. Econometric analysis is often used to identify and quantify the

impact of allegedly illegal acts and distinguish that impact from effects

caused by other factors.

To illustrate when and how econometric analysis can be useful, this

chapter considers estimating damages in an alleged price-fixing case in

which defendants are alleged to have fixed the sale price of the product

in question, i.e., a sell-side price-fixing conspiracy. Assuming there has

been an agreement to raise prices, the basic questions relating to the

calculation of damages are (a) whether the alleged conspiracy affected

prices (the fact of injury) and, if so, (b) the extent to which prices have

increased as a result of the alleged conspiracy (the amount of injury).

1

Because prices prevailing in the market likely were affected by a wide

variety of other demand and supply factors unrelated to the alleged

conspiracy, isolating and measuring the effect of the alleged conspiracy

on price requires properly accounting for these other factors. Otherwise,

one might confuse the effects of one or more of these other factors with

the effects of the alleged conspiracy, thus overestimating or

underestimating damages attributable to the alleged anticompetitive

conduct. The goal is to isolate the impact of the conspiracy on prices (the

overcharge), and then multiply the overcharge by the quantity of sales to

calculate damages recoverable by the purchasers.

1

. See John E. Lopatka, Antitrust Injur y and Causa tion, in 3 ISSUES IN

COMPETITION L. & POL’Y 2299, 2311 (ABA Section of Antitrust Law

2008); Theon van Dijk & Frank Verboven, Qua ntification of Damages, in

3 ISSUES IN COMPETITION L. & POL’Y 2331, 2 334 (ABA Section of

Antitrust Law 2008).

302 Econometrics

A. Non-Econometric Approaches to Damages

One approach to estimate the overcharge could be to perform a

simple comparison of prices in the market before and after an allegedly

anticompetitive event (i.e., the alleged agreement to fix prices) to

determine if prices increased after the alleged agreement.

2

The

underlying premise of this approach is that the market has not changed

substantially over the relevant time period, except as a result of the

allegedly illegal behavior.

Figure 1 below provides a hypothetical example in which prices are

substantially higher in the period after the alleged price-fixing agreement

than they were before the agreement came into force. The dots show

actual prices over the entire period. The horizontal line shows the

average price before the alleged agreement, which extends into the

period after the alleged agreement. Absent any substantial changes in

supply and demand in the market, an economist might use the average

price in the pre-conspiracy period to predict prices during the alleged

damages period, but for the alleged conspiracy. The difference in average

prices between the pre-conspiracy and post-conspiracy periods is

illustrated in Figure 1 as the difference between the dots and solid line

after the date of the alleged agreement (shown as the vertical line). The

difference between these actual and estimated prices “but for” the alleged

conspiracy multiplied by the units of sales in the conspiracy period might

be used to measure damages,

3

so long as (a) there are no significant

changes in the market other than the alleged illegal behavior and (b) the

time period of the alleged anticompetitive conduct has been identified

correctly.

2

. See detailed discussion of before and “during” analyses below.

3

. In addition to estimating the but-for price in this conspiracy example,

economic analysis also predicts that the but-for quantity could be larger

than actual quantity because of the lower but-for price. However, most

conspiracy cases focus on overcharges as the measure of the da mages.

Accordingly, this example focuses on overcharges.

Applying Econometrics to Estimate Damages 303

Figure 1. Average Price Approach to Price Overcharge

There are, however, limitations to this approach.

4

In particular, other

important external influences on prices often have changed during the

time period of the alleged anticompetitive actions and thus may explain

the price increase, at least to some degree. The simple approach

described above would mistakenly attribute the effects of these external

influences to the alleged conspiracy. If so, this approach would not be

reliable for estimating prices but for the alleged conspiracy.

Several methods can be used to control for other influences. For

example, using the example above but assuming that a significant cost

increase occurred at or about the time of the alleged price-fixing

agreement, which similarly affected all firms in the market, a more

complete analysis than just comparing prices before and after the alleged

4

. For exa mple, d etermining the end of the alleged conspiracy to coincide

with a decline in price can cause a bias to finding damages and

suggesting a conspiracy. Dennis Carlton and Gregory Leonard have

suggested an app roach to address this problem. See Dennis Carlton &

Gregory K. Leonard, Correcting the Bias When Damage Periods Are

Chosen to Coincide With Price Declines, 2004 COLUM. BUS. L. REV. 304

(2004).