Securities Class Action Economist & Event Study Expert

A securities class action turns on a short list of empirical questions: whether the security traded in an efficient market, whether the alleged misstatements affected its price, how much of the decline on each corrective-disclosure day the alleged fraud explains rather than the market, the industry or other company news, and what that means per share and across the class. We answer those questions with event studies and class-wide damages models built so that every input, from the choice of index to the length of the estimation window, is stated and can be rerun by the other side. The work covers equity, debt, options and other derivatives, and commodity markets, and we are retained on both sides of the caption.

What we analyze

• Market efficiency for the security and class period, tested under the Cammer and Krogman factors, including the cause-and-effect analysis of how the price responded to unexpected company-specific news.
• Price impact of the alleged misstatements and of the disclosures said to correct them.
• Loss causation and corrective disclosures: how much of each decline is attributable to corrective information and how much to market, industry, earnings and other confounding information released at the same time.
• Materiality and economic significance: whether a price reaction is distinguishable from the security’s ordinary variation, and how large it is in dollars and in percent.
• Inflation and class-wide damages: the inflation ribbon across the class period, per-share damages by purchase and sale date, the PSLRA 90-day lookback, aggregate damages with the trading-model assumptions stated, and plans of allocation.
• Section 11 and Section 12 damages: the statutory measure for registered offerings, negative causation and tracing.
• Review of the opposing expert’s event study, replicated from its data and workpapers and tested one choice at a time.

Event studies: how we build one

An event study asks whether a security’s price moved on a given day by more than the market and its industry would explain, and whether that movement is tied to the information at issue. It is the tool behind the market-efficiency showing, the price-impact question, loss causation and the inflation ribbon, and each step of it is specific enough to be tested line by line.

• Estimation window. A period, commonly 120 to 250 trading days, used to measure how the security normally moves with the market and its industry. It should be free of the events being tested and, where possible, of the alleged fraud, so the benchmark is not contaminated by the conduct at issue. Whether it sits before the class period, after it or around each event is a stated choice, and the report shows whether the conclusion survives a different window.
• Expected returns. A regression of the security’s daily return on the return of a market index and, in most cases, an industry index or a peer portfolio. The estimated coefficients give the return the security would have been expected to earn on any day, given what the market and the sector did. The choice of indices matters: a peer index that includes the defendant, or that one company dominates, biases the benchmark, and the report says which indices were used and why.
• Abnormal returns. The actual return on the event day minus the expected return. Multiplied by the prior day’s closing price, it gives the abnormal dollar change per share, which is the input to the damages model.
• Statistical significance. The abnormal return is compared with the security’s normal day-to-day variation over the estimation window. By convention a move is called statistically significant when it would occur by chance less than five percent of the time, which corresponds to a test statistic of roughly two in absolute value. Where volatility changed during the period, the test has to use the volatility of the relevant period or it misstates the evidence. Event dates are fixed from the content and timing of the news before the price data are examined.
• Confounding information. Earnings, guidance, analyst actions, management changes and macro news that reached the market the same day have to be separated from the corrective disclosure, or the analysis misstates what the alleged misstatement did to the price. How we do that is set out under loss causation below.

The method is set out in more detail in our note on event study methodology in securities litigation, and a worked illustration on public data is in our event study of the May 2025 Compass Diversified disclosure.

Market efficiency and class certification

Reliance is presumed class-wide under Basic Inc. v. Levinson, 485 U.S. 224 (1988), when the security traded in an efficient market, so market efficiency is the first empirical question at certification. Courts assess it under the factors in Cammer v. Bloom, 711 F. Supp. 1264 (D.N.J. 1989): average weekly trading volume, analyst coverage, the presence of market makers and arbitrageurs, eligibility to file a Form S-3, and a cause-and-effect relationship between unexpected corporate events and price movements; Krogman v. Sterritt, 202 F.R.D. 467 (N.D. Tex. 2001), adds market capitalization, bid-ask spread and public float.

The first four Cammer factors and the Krogman factors are documented from trading, filing and ownership data. The fifth factor is the one that requires an economist, and it is the one most contested motions turn on. We run the event study across the class period’s earnings announcements and other unexpected news days and compare the frequency and size of significant price reactions on those days with the reactions on days without company-specific news. A security that responds promptly to new information on news days and not otherwise is behaving as the presumption assumes; a security that does neither, or that drifts for days after an announcement, is not. Efficiency is a property of a particular security in a particular period, and the analysis treats it that way, which matters for thinly traded stocks, for debt and for securities whose trading changed during the class period.

Certification also asks whether damages can be measured class-wide by a method tied to the liability theory (Comcast Corp. v. Behrend, 569 U.S. 27 (2013)). Courts of appeals have recently vacated or reversed certification where the plaintiffs’ damages methodology was described only in general terms and the choice of method deferred (In re FirstEnergy Corp. Securities Litigation (6th Cir. 2025); Rhode Island Office of the General Treasurer v. The Boeing Co., No. 25-1492 (4th Cir. July 20, 2026)); the Fourth Circuit’s phrase was that a menu of options to be chosen later is not a methodology. What we offer at certification is a specified model: the corrective disclosures it uses, how inflation is carried back across the class period, how confounding information is handled, and how per-share damages are computed for each purchase and sale date.

Price impact

Since Halliburton Co. v. Erica P. John Fund, Inc., 573 U.S. 258 (2014), defendants may rebut the presumption of reliance at certification by showing that the alleged misstatements had no impact on the price, and under Goldman Sachs Group, Inc. v. Arkansas Teacher Retirement System, 594 U.S. 113 (2021), they bear the burden of persuasion on that question. Price impact is an empirical question about the security, and the economist’s work on it has two ends.

At the front end, the question is whether the price moved when each alleged misstatement was made. Often it did not, because the statement confirmed what the market already believed, which is why many cases proceed on a price-maintenance theory: the misstatement kept an already inflated price from falling. On that theory the evidence shifts to the back end, the corrective-disclosure days, and the analysis asks whether the price fell when the truth emerged and whether that fall is attributable to the correction of the specific statement. After Goldman, courts also weigh the fit between a generic front-end statement and a specific back-end disclosure, and the Second Circuit decertified the Goldman class on that ground (77 F.4th 74 (2d Cir. 2023)). Our analysis therefore pairs each alleged misstatement with the disclosure said to correct it, tests the price reaction at both ends, and states what the pairing does and does not support. The same work, from the other direction, is the rebuttal of a price-impact defense.

Loss causation and corrective disclosures

Under Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336 (2005), the recoverable loss is the decline caused by the revelation of the alleged misstatement, not the decline caused by anything else that happened to the company or the market. The economist’s task is to isolate that decline, disclosure by disclosure. How that showing differs from price impact at certification, in burden, timing and what the event study must establish, is set out in loss causation vs. price impact.

• Evaluating the alleged corrective disclosures. For each disclosure date in the complaint we identify what information actually reached the market, when it reached it relative to trading hours, whether it was new, and whether and how it related to the alleged misstatements. Whether a disclosure corrected the alleged misstatement is a legal question, and the report presents the record on content and timing rather than a conclusion on it. Partial disclosures, disclosures spread over several days and information that leaked before the formal announcement are identified, because they change where inflation leaves the price.
• Market and industry movement. The market model removes the part of the day’s return explained by the market and the industry. On a day when the sector fell, the abnormal return is smaller than the raw decline, and only the abnormal return is carried into the damages analysis.
• Earnings and other company-specific information. A corrective disclosure rarely arrives alone. Where earnings, guidance, a management change or an analyst action landed the same day, the abnormal return has to be apportioned. The approaches include intraday price analysis where the items were released at different times, analyst and press commentary on what the market reacted to, comparison of the earnings surprise with the market’s typical reaction to surprises of that size, and bounding the fraud-related portion from above and below. A report that attributes the entire decline to the alleged fraud on a day with a large earnings miss, without this analysis, has skipped the step.
• What the report says. For each disclosure day: the raw return, the expected return, the abnormal return and its significance, the confounding information identified, the apportionment and its basis, and the fraud-related dollar decline per share that results. Those figures are the building blocks of the inflation ribbon.

Materiality and economic significance

Whether information was material is a legal question about the reasonable investor. The price evidence bears on it in two ways. A statistically significant reaction to a disclosure is evidence that the information mattered to investors, although statistical significance is not a legal prerequisite for materiality (Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27 (2011)). And economic significance is distinct from statistical significance: a reaction can be statistically significant and small, or large and, for a volatile security, within the range of ordinary noise. We report both, the test statistic and the magnitude in percent and in dollars per share, together with the security’s normal daily variation, so the reader can weigh them rather than take one number for the other.

Damages: inflation, inflation ribbons and class-wide damages

Damages under Rule 10b-5 are out-of-pocket damages: the difference between the price paid and the price that would have prevailed absent the misstatements, for each share, on each day. The inflation ribbon is that difference across the class period. It is built backward from the corrective disclosures: the fraud-related abnormal dollar decline on each disclosure day is the inflation removed that day, and the inflation on any earlier day is the sum of what remained to be removed. Whether inflation is carried back as a constant dollar amount or as a constant percentage of price, how partial disclosures step the ribbon down, and how intervening company news is treated are choices the report states and tests, because they change who in the class has damages and how much. The full method, from the ribbon to the aggregate figure and the plan of allocation, is in calculating class-wide securities damages.

Per-share damages for a class member are the inflation at purchase less the inflation at sale, or less zero for shares held through the last disclosure, subject to the limit in 15 U.S.C. § 78u-4(e): damages may not exceed the difference between the purchase price and the mean trading price over the 90 days after the corrective information was disseminated, or through the date of sale for shares sold within that window. Shares bought and sold before any corrective disclosure carry no damages under this measure, and the report says so.

Aggregate class-wide damages require, in addition, an estimate of how many damaged shares were held and traded by class members across the period. The trading models that produce that estimate rest on assumptions about turnover and trader behavior, and we present them with the assumptions stated and bounded rather than as a single figure. The same inputs produce a plan of allocation at settlement: a recognized-loss schedule by purchase and sale date, with the lookback applied, and a loss estimate for the largest-financial-interest showing on a lead-plaintiff motion (15 U.S.C. § 78u-4(a)(3)(B)).

Section 11 and Section 12. Claims under Section 11 of the Securities Act (15 U.S.C. § 77k) for misstatements in a registration statement carry a statutory measure: the difference between the amount paid, not exceeding the offering price, and the value at the time of suit, or the price at which the security was sold before suit, or the price at which it was sold after suit but before judgment if that is less. Scienter is not an element and reliance generally is not, and loss causation enters as the defendant’s negative-causation defense, which reduces damages by the part of the decline shown to reflect something other than the misstatement. That proof is an event study run in reverse, and the economic work is the same whether we are building the decline attributable to the registration statement or isolating the portion that is not. Section 11 standing requires shares traceable to the registration statement (Slack Technologies, LLC v. Pirani, 598 U.S. 759 (2023)), which after a direct listing or lock-up expiry is a question about share provenance and trading data. Section 12(a)(2) claims (15 U.S.C. § 77l) provide rescission or rescissory damages, subject to the loss-causation defense in Section 12(b).

Where the analysis attaches in a Rule 10b-5 case

• Before the complaint. A preliminary event study on the alleged corrective-disclosure days, from public price and volume data, shows whether there is a statistically significant decline to build a loss-causation theory on and how much competing news there was on those days, and a first loss estimate sizes the case for the lead-plaintiff motion.
• Class certification. Market efficiency, price impact and a committed, liability-matched damages methodology, as above.
• Merits. Loss causation and materiality, disclosure by disclosure.
• Damages and settlement. The inflation ribbon, per-share and aggregate damages, mediation ranges broken out by disclosure and by assumption, and the plan of allocation.

For a stage-by-stage account written for counsel new to these cases, see Rule 10b-5 class actions: the economist’s role at each stage and calculating damages in Rule 10b-5 cases.

Reviewing the opposing expert’s event study

Most disputes between securities experts concern the choices behind the event study rather than its arithmetic: the indices and the estimation window; whether event dates were selected with knowledge of the price reaction; how multi-day windows, after-hours releases and trading halts were handled; whether significance was tested against the security’s volatility in the relevant period; how confounding news was apportioned; and whether the inflation ribbon reflects partial corrective disclosures or carries the whole final decline back to the first day of the class period. We replicate the opposing analysis from its data and workpapers, identify the choices that drive its conclusion, and show what happens when each one is changed. If you have received a report, send it to us for review.

Who leads the work

Securities engagements are led by two economists. Cristiano Manfre, Ph.D., an affiliated economist (Ph.D. and M.A. in Economics, Claremont Graduate University; MBA in Finance, Drucker School of Management), holds FINRA Series 7, 24, 28, 63 and 79 licenses, led the corporate valuation practice in Marshall & Stevens’ Los Angeles office, and teaches mergers and acquisitions and business valuation at Pepperdine University’s Graziadio Business School.

Roman Garagulagian, Ph.D., the firm’s principal (Ph.D. in Economics, Claremont Graduate University; Lecturer in Finance at Cal Poly Pomona since 2014), leads the econometric work (event-study specification, significance testing and the damages model) and the options and derivatives questions: damages for options traders in the class, options-market evidence offered on market efficiency, and options trading in insider-trading and manipulation claims. His article “Strike Salience and Liquidity Quality in U.S. Equity Options Markets” is forthcoming in The Journal of Derivatives (Spring 2027). He has testified as a damages expert in federal and state matters since 2017.

The firm’s economists support the data assembly, modeling and replication work under their direction. Every engagement is led by the economist responsible for the opinion. Analysts may support the work, but the designated expert owns the methodology, assumptions, analysis, report and testimony. We do not designate one expert and hand the file to another.

What we need to start

1. Party names only, for a conflict check. Nothing privileged.
2. A 20 to 30 minute scoping call on the security, the class period, the alleged misstatements and corrective disclosures, and the deadline.
3. Once engaged: the operative complaint; the security identifiers and, for debt or derivatives, the terms; the offering documents and SEC filings for the period; any analyst reports and news already collected; trading data where the matter is past the public-data stage; and any expert reports already served, with their workpapers. Much of the market data is public, which is why a preliminary read on price impact is often possible before the engagement is scoped in full.

You receive a written scope and phase budget before any engagement letter. If the price data will not carry the opinion you need, we tell you at the scoping stage.

If you have a matter now, start with a conflict check using party names only. If the other side has already served an event study or damages report, send us the report for review. When the file is ready, upload the case files securely.

Frequently asked questions

Can you tell early whether an event study will support the claim?

Often, yes. Price and volume data around the alleged disclosure dates are public, and a preliminary analysis shows whether there is a statistically significant move to build on and how much same-day news competes with the corrective disclosure. That is a short piece of work and it is better done before positions harden.

Do you handle class certification as well as damages?

Yes. At certification the questions are market efficiency, price impact and whether damages can be measured class-wide by a method tied to the liability theory. All three are empirical questions about the specific security, and the certification report commits to the model that will be used at the merits.

What is the difference between statistical and economic significance?

Statistical significance says whether a price move is distinguishable from the security’s ordinary variation. Economic significance is the size of the move in percent and in dollars per share. A report should give both; a small but significant move and a large but insignificant one call for different arguments.

What is the difference between Rule 10b-5 and Section 11 damages?

Rule 10b-5 damages are built from the inflation per share across the class period and limited by the PSLRA lookback. Section 11 damages are set by statute from the price paid, capped at the offering price, and the value at suit or sale, with the defendant able to reduce them by proving negative causation. The event-study work overlaps; the arithmetic that follows does not.

Can you work on thinly traded securities, debt or options?

Yes. Market efficiency and price impact are harder to establish for a thinly traded security, and the analysis has to say so plainly rather than force the result. Debt instruments and options require their own benchmarks and pricing models, and the approach is set out in the scope before the work begins.

Can you work as a rebuttal expert?

Yes, and a meaningful share of securities work is rebuttal. Tell us which role you need at the conflict-check stage.

Do you prepare plans of allocation and loss estimates for lead-plaintiff motions?

Yes. A loss estimate for the largest-financial-interest showing under 15 U.S.C. § 78u-4(a)(3)(B) uses the same inputs as the damages model, and a plan of allocation at settlement is the recognized-loss schedule those inputs produce.

What are your rates?

We send the rate schedule and a full CV with the proposal after the scoping call, so you are pricing against a defined scope.