Constructing a Beta-Neutral Insider Portfolio

Note on figures cited below. Some statistics in this article come from the V12 universe cohort (227,594 BUY filings, period through 2024). Since 2025-01-01 the live out-of-sample BUY universe (28 markets) shows a 52.96 percent win rate and a +8.59 percent mean 90-day return on 20,688 filings. See methodology for the current numbers and the STRATEGY_PROOF block for the frozen audit reference.

A clean insider signal can survive contact with the market, but only if you stop asking it to carry the index on its back.

Insider trading strategies have an awkward habit. They look brilliant in a chart until the market falls 20%, at which point one discovers that a portfolio of “informative” insider buys is still, regrettably, a portfolio of equities.

That is the case for beta-neutral construction. If the signal is meant to capture firm-specific information embedded in directors’ and executives’ trades, then broad market exposure is noise, occasionally expensive noise. A long-only insider portfolio mixes two bets: first, that insiders know something useful about their own firms, and second, that equities in general will cooperate. The second bet is optional.

This article sets out a practical construction recipe for a beta-neutral insider portfolio, long the insider signal and short an index future or ETF. The aim is not to produce a perfect market vacuum. It is to reduce drawdowns, improve attribution, and make the strategy’s failures more honest.

Why hedge an insider strategy at all?

The short answer is that insider portfolios are usually long beta whether the manager admits it or not.

A portfolio built from insider purchases tends to lean toward smaller companies, cyclical businesses, and firms that have recently sold off enough to look attractive to management. Those characteristics can be useful. They also tend to come with above-market sensitivity in risk-off periods. If the market drops sharply, many insider-heavy portfolios draw down not because the insider signal failed, but because the portfolio was simply net long equities with a distinct small-cap flavour.

Beta-neutrality addresses a narrow but important question: after removing broad market exposure, is there still a stock-selection effect left?

Alpha attribution, with fewer excuses

A long-only insider strategy can post handsome returns in a bull market and still teach us very little. If the benchmark rises 25% and the portfolio rises 30%, the excess return may be signal, sector bias, factor exposure, or luck wearing a tie. A hedge does not solve every attribution problem, but it removes the largest and laziest one.

This matters especially for insider data because the academic literature generally finds that insider purchases, more than insider sales, contain predictive information, but the effect is uneven across firm size, information environment, and market regime. Once one knows that the signal is conditional, not universal, isolating it from market direction becomes less a luxury and more basic hygiene.

Drawdown reduction is the real sales pitch

The phrase “beta-neutral” often gets sold as if it were a return enhancer. Usually it is a path improver. Hedging broad market exposure can reduce peak-to-trough losses in market sell-offs, and that can be strategically valuable even if headline returns end up lower in runaway bull markets.

That trade-off is not glamorous. It is, however, useful. Lower drawdowns can mean less capital flight, fewer forced de-risking decisions, and a better chance that the strategy survives long enough for the insider signal to matter.

The construction recipe

The recipe is straightforward on paper. The irritation lies in the details, which is usually where returns go to die.

Step 1: Define the insider signal with some restraint

Start by deciding what counts as a meaningful insider event. The literature and many practitioners place more weight on open-market purchases than on sales, because sales are often driven by diversification, tax, or liquidity needs. Purchases are rarer and more likely to reflect conviction.

A basic signal definition might include:

• Open-market purchases only.
• Directors and senior executives only, or a broader set of reportable insiders depending on jurisdiction.
• Minimum transaction size, either in currency terms or relative to the insider’s prior holdings.
• Optional clustering rules, such as multiple insiders buying the same issuer within a short window.
• Exclusions for automatic trading plans where identifiable, option exercises without net open-market buying, and obvious compensation-related transactions.

This is where regulation matters. In the United States, many transactions are reported on Form 4 within two business days. In the EU under MAR Article 19, persons discharging managerial responsibilities, PDMRs, and their closely associated persons must notify transactions within three business days, with the issuer then disclosing. Those lags are short, but not trivial. A strategy must trade on the date information becomes public, not the date the insider traded. Otherwise one is backtesting clairvoyance, which remains difficult to implement.

Step 2: Aggregate signals into a long book

Once events are defined, turn them into positions. There are several common approaches:

Equal-weight by event

Each qualifying insider purchase generates a fixed position size. This avoids letting mega-caps dominate the book and often suits a signal that is strongest in smaller names. It also increases turnover and transaction-cost sensitivity.

Value-weight by signal strength

Position sizes scale with a measure such as transaction value, insider rank, number of co-buyers, or purchase size relative to salary or prior holdings. This can improve economic intuition but invites overfitting if too many knobs are added.

Issuer-level aggregation

If multiple insiders buy the same stock within a window, combine them into one issuer score rather than stacking separate positions. This reduces accidental concentration in names with unusually active boards.

Holding periods are another design choice. Many insider anomalies decay over weeks to months rather than days. A common recipe is to hold each signal for one to six months unless stopped out by a contrary event or portfolio rule. The right answer depends on the jurisdiction, liquidity universe, and cost budget.

Step 3: Estimate portfolio beta

Now the hedge. Compute the long portfolio’s beta to the benchmark you intend to short. This can be done using rolling regressions of portfolio returns on benchmark returns, or by aggregating constituent betas weighted by current position sizes.

The first method captures realised portfolio behaviour. The second is faster and more transparent, but depends on stock-level beta estimates that can be noisy, especially in small caps.

A practical compromise is:

1. Build the long portfolio.
2. Estimate each constituent’s beta to the chosen benchmark using a rolling window, for example 60 to 252 trading days.
3. Aggregate to a portfolio beta using current weights.
4. Cross-check against realised portfolio beta from a shorter rolling regression.

If the two estimates disagree wildly, the portfolio is telling you something, usually that your beta model is unstable or your constituents do not trade cleanly enough for elegant statistics.

Step 4: Short the benchmark in proportion to beta

If the long book has a market value of 100 and an estimated beta of 1.15 to the benchmark, short 115 of benchmark notional to target zero beta. That is the core mechanic.

The benchmark choice should match the investable universe as closely as possible:

• US large-cap insider universe: S&P 500 future or ETF may be reasonable.
• Broader US universe with small-cap tilt: Russell 2000 or Russell 3000 proxy may fit better.
• Pan-European book: STOXX Europe 600 or a country mix.
• Single-country strategy: local broad market index.

There is no prize for hedging French small-cap industrials with the Nasdaq 100. The resulting chart may be exciting, but the attribution will be comic.

Step 5: Rebalance on a schedule, not on emotion

Beta drifts as prices move, names enter and leave the portfolio, and market regimes change. Rebalance the hedge at a set frequency, daily if capacity and infrastructure allow, weekly if the strategy is lower turnover and cost-sensitive.

A common mistake is to rebalance only the long book while leaving the short leg stale. The result is a strategy that starts beta-neutral and ends up expressing a market view by accident.

A concrete implementation framework

Theory is tidy. Portfolio plumbing is not. Here is a workable institutional recipe.

Universe selection

Begin with a liquid universe where insider disclosures are timely and executable. Apply minimum thresholds for market capitalisation and average daily value traded. If the signal is strongest in illiquid microcaps but the portfolio cannot trade them without becoming the market, the backtest is a museum exhibit.

For cross-market strategies, harmonise disclosure fields carefully. “Insider” is not a universal category in practice. Filing formats, role labels, transaction codes, and amendment processes differ materially between regulators.

Signal scoring

A robust score often combines a few intuitive features rather than a dozen fragile ones:

• Role weight: CEO and CFO purchases may carry more information than non-executive directors, though evidence varies by market.
• Trade size: Larger purchases relative to compensation or existing ownership tend to be more credible than symbolic buys.
• Cluster effect: Multiple insiders buying in a short interval is often treated as a stronger signal than a single buyer.
• Recent returns: Some strategies prefer insider buys after sharp declines, on the theory that insiders exploit mispricing. Others avoid catching falling knives. Both camps can produce attractive slides.

Keep the score parsimonious. Every extra parameter improves historical fit until it does not.

Position sizing

Position sizing is where risk control becomes visible. A sensible framework includes:

• Maximum weight per name.
• Sector caps.
• Liquidity-aware sizing.
• Country caps for international books.
• Optional volatility scaling.

For beta-neutral portfolios, it is often useful to size the long book first under these constraints, then compute the hedge notional from the resulting portfolio beta. Do not set the short leg independently and hope the arithmetic works itself out.

Hedge instrument choice

The benchmark short can be implemented with futures, ETFs, or swaps.

• Futures are usually capital-efficient and liquid, but require operational capability and roll management.
• ETFs are simple and accessible, but can introduce borrow costs, tracking differences, and tax wrinkles.
• Swaps can be flexible for institutions, but they add counterparty and financing considerations.

For many managers, index futures are the cleanest tool. For many retail implementations, they are not available or not suitable, which is one reason public discussion of “beta-neutral” strategies often quietly assumes professional infrastructure.

Where drawdown reduction actually comes from

A hedge helps in three main ways, and none are mysterious.

It removes the obvious market leg

If the market falls 15% and the long insider book has a beta above one, a large share of the drawdown is broad market exposure, not signal failure. Shorting the benchmark offsets that component, at least to the extent beta was estimated correctly and remains stable.

It reduces forced correlation in stress

During market stress, correlations rise. Even stocks with idiosyncratic positive information often decline alongside the tape. A benchmark short cushions this regime effect. It will not save every position, but it can stop the portfolio from behaving like a leveraged version of the market at the worst possible time.

It improves behavioural survivability

This is not a line item in a factor model, but it matters. Strategies with smaller drawdowns are easier to hold, easier to explain, and less likely to be abandoned near the trough. The investor who can stick with the process has an edge over the investor who liquidates after discovering that “long-term horizon” meant about eleven unpleasant trading days.

The trade-off: hedging can dilute upside

There is no free lunch, only differently priced catering. In strong bull markets, a beta-neutral insider portfolio will often lag its long-only cousin because the hedge offsets some of the market tailwind. That is not a bug. It is the bill for lower market dependence.

The relevant question is whether the resulting return stream is superior on a risk-adjusted basis and whether it better reflects the underlying information signal. If the answer is yes, lower raw return in some periods may be entirely acceptable.

The main failure modes

Most insider strategies do not fail because the concept is absurd. They fail because implementation is casual.

Disclosure lag and event timing

The first and most common mistake is timing. Insiders trade before the market knows. You do not. The strategy can only enter after public disclosure, which may be one to three business days later depending on market and filing mechanics. Event studies that anchor on transaction date rather than filing date flatter the signal.

Small-cap and liquidity bias

Insider effects are often stronger in smaller, less-followed firms. That is interesting and inconvenient. The stronger the signal appears in illiquid names, the more likely transaction costs and market impact will eat it. A beta hedge does nothing to solve this. It can even create a false sense of safety, because market beta is neutralised while liquidity risk remains fully intact.

Beta instability

Beta is not a law of nature. It changes with market regime, sector composition, and stock-specific events. A portfolio that was 0.9 beta last month may be 1.3 in a panic. Shorting a fixed benchmark notional based on stale estimates can leave the portfolio under- or over-hedged precisely when it matters most.

Factor contamination

An insider portfolio can carry substantial exposure to size, value, momentum reversal, quality, or leverage. A simple market hedge removes only the first-order index effect. If the manager wants purer stock-specific alpha, further factor hedging may be warranted. That moves the strategy from “beta-neutral” toward “multi-factor neutral”, which is more elegant and more operationally demanding.

Short-leg frictions

The short index leg is not frictionless. Futures require margin and rolling. ETFs may involve borrow costs. Swaps have financing spreads. During stress, basis can move in unhelpful ways. These costs are usually manageable, but they must be counted. A backtest that treats the hedge as free is not a backtest. It is fan fiction.

What the literature suggests, and what it does not

The academic evidence on insider trading is broad, but a few themes are durable.

Insider purchases tend to be more informative than sales

This is one of the more stable findings. Purchases are less frequent and more discretionary. Sales are noisy because insiders sell for many reasons unrelated to expected underperformance. A long-biased strategy built around purchases therefore has a stronger conceptual basis than a short strategy built around sales alone.

Information effects are stronger where information asymmetry is higher

Smaller firms, less-covered firms, and firms with greater uncertainty often show stronger post-disclosure effects. Again, useful and inconvenient. The signal lives where execution gets harder.

Public disclosure does not eliminate all predictive content

Markets react to insider filings, but not always fully or instantly. That is the opening a systematic strategy seeks to exploit. The persistence of this effect likely reflects processing frictions, heterogeneous attention, and the fact that insider trades are informative but noisy. If they were perfectly informative and perfectly interpreted, there would be no strategy, only a footnote.

Beta-neutrality is a portfolio overlay, not a separate anomaly

The literature on insider trading mostly studies abnormal returns relative to benchmarks and factor models. Translating that into a live beta-neutral portfolio is a practical overlay. It does not create information where none exists. It simply aims to preserve the information content while reducing contamination from broad market moves.

That distinction matters. If the insider signal is weak after realistic costs and delays, hedging will not rescue it. It may produce a smoother path to mediocrity, which is still mediocrity.

A minimal working example

Suppose a manager runs a US insider-buy strategy with 40 names, equal-weighted, refreshed weekly, and each signal held for 60 trading days. On Monday’s rebalance, the long book totals 100 units of capital. Using 126 trading days of daily returns, the manager estimates the portfolio beta to the Russell 2000 at 1.10.

The manager then shorts 110 units notional of Russell 2000 futures.

A week later, market moves and constituent changes push estimated portfolio beta to 0.95. The short is resized to 95 units. If the market falls sharply during that week, the hedge should offset much of the broad-market loss, leaving the portfolio’s return more dependent on whether the insider-selected names outperform the benchmark on a relative basis.

This is not exotic. It is just disciplined.

Variants worth considering

Sector-matched hedging

If the insider book is heavily tilted toward one or two sectors, a broad index hedge may be blunt. One can hedge with a combination of sector and market indices. This reduces residual systematic exposure but increases complexity and turnover.

Dollar-neutral rather than beta-neutral

Some managers simply match long and short notionals one-for-one. That is easier, but not the same thing. A long book of high-beta small caps and a short in a broad index can still have substantial net market exposure even if dollar-neutral.

Dynamic hedge bands

Instead of rebalancing continuously, some managers allow beta to drift within bands, for example ±0.1, before resizing the hedge. This reduces trading costs on the short leg at the expense of more variable market exposure.

What to watch in live deployment

A live beta-neutral insider strategy should be monitored with a short checklist that is more operational than philosophical.

Daily diagnostics

Track:

• Net and gross exposure.
• Estimated beta and realised beta.
• Sector and size exposures.
• Liquidity concentration.
• Hedge slippage and basis.
• Contribution from long selection versus hedge.

If realised beta persistently differs from target beta, the hedge process needs repair. If the strategy’s returns are mostly explained by size or value rebounds, the insider signal may be receiving undue credit.

Event quality control

Insider data is messy. Amendments, duplicate filings, role misclassifications, and transaction-code errors are common enough to matter. A production strategy needs validation rules for:

• amended filings,
• linked insiders,
• transaction type mapping,
• corporate actions,
• stale disclosures.

The phrase “garbage in, garbage out” survives because it remains underappreciated by people with attractive backtests.

Capacity limits

Capacity is usually lower than the cleanest charts imply. If the signal is strongest in smaller names and the portfolio must enter after public disclosure, crowding and impact can rise quickly. A beta hedge does not increase capacity in the long book. It merely makes the ride less dependent on the market’s mood.

The practical verdict

A beta-neutral insider portfolio is not a trick. It is a sensible way to ask a better question of the data. Instead of “did insider buys make money in a rising market?”, the question becomes “did stocks bought by insiders outperform after controlling for the market?”

That shift improves both research and live risk management. The expected reward is not necessarily higher absolute return. It is cleaner attribution and, in many regimes, lower drawdown. For allocators and managers who care about the path as much as the destination, that is not a minor benefit.

The next concrete step is simple: run the same insider strategy twice, once long-only and once beta-neutral using filing publication dates and full trading frictions, then compare max drawdown, recovery time, and realised beta. The open question is where to stop hedging, market only, or market plus sector and style factors, before the overlay starts neutralising the very inefficiencies the strategy was hired to exploit.

Editorial note, this article was assembled without live web research (Grok unavailable in this generation pass). Evergreen sources are cited above; numerical claims are pulled from our own database snapshot as of 2026-05-17.

Insiders Trades backtest cohort

Our analysis of the beta-neutral insider portfolio strategy reveals significant insights from a robust dataset of 220,691 priced trades across 17 markets. The mean 90-day return for this cohort stands at -1.70%, with a win rate of 41.82%. This performance aligns with the broader global universe context, which shows an aggregate mean 90-day return of -1.70% and an annualized Sharpe ratio of -0.08. These figures underscore the challenges faced by investors in achieving positive returns in a market characterized by insider trading dynamics. The findings suggest that while beta-neutral strategies can mitigate risk, they may not guarantee outperformance in the current market environment. It is essential to consider these statistics in the context of the sample size and time horizon analyzed.

Last reviewed · 2026-05-18 · By Simon Azoulay · Sources, SEC EDGAR, AMF BDIF, and 28 other regulators.