Ethical Quant Trading: Regulations & Market Manipulation › Ethical Use of Data

The explosion of alternative data, which we explored in the chapter on alternative data and NLP in quant strategies, raises ethical questions that go beyond legal compliance. Just because data is technically available does not mean it is ethically appropriate to use. The distinction between legality and ethics is critical here: legal standards define the floor of acceptable behavior, while ethical standards aspire to a higher ceiling.

Many alternative data sources derive value from aggregating individual behavior. Credit card transaction data, mobile phone location data, social media activity, and web browsing patterns can all generate alpha signals. But the individuals whose data is being analyzed typically did not consent to their behavior being used to inform trading decisions. The aggregation itself does not eliminate the ethical concern: even when individual records are combined into aggregate statistics, the value of the data derives from monitoring individuals' private behavior.

Key ethical questions include:

• Consent and expectation: Did the data subjects reasonably expect their data to be used for financial trading? Even if a privacy policy technically permits it, does that reflect meaningful consent? A consumer who agrees to a credit card company's terms of service likely does not envision that their purchase history will be sold to a hedge fund and used to predict a retailer's quarterly earnings.
• Re-identification risk: Even "anonymized" datasets can sometimes be re-identified when combined with other data sources. A dataset showing purchase patterns at a specific store in a small town may effectively identify individuals. Research in data science has repeatedly demonstrated that supposedly anonymous datasets can be de-anonymized through linkage with other available information, raising the question of whether true anonymization is even achievable for many types of behavioral data.
• Data provenance: Was the data collected lawfully? Web scraping may violate terms of service. Purchasing data from vendors who obtained it through deceptive practices can expose you to legal and reputational risk. Your responsibility does not end at the point of purchase: if you know or should know that a data vendor's collection methods are ethically questionable, using that data implicates you in the ethical breach.

Expert networks, which connect analysts with industry experts, have been the source of numerous insider trading cases. The SAC Capital (now Point72) and Galleon Group prosecutions revealed how seemingly legitimate expert consultations can cross the line into tip-based trading. If you use expert networks or any form of primary research that involves speaking with corporate insiders, you need compliance protocols that document that no MNPI was exchanged. These protocols typically include pre-clearance of expert contacts, monitoring of conversations, and written records of the topics discussed and the information obtained.

A growing number of investors and regulators expect quantitative strategies to consider ESG factors, or at minimum to disclose their approach. Even if you are agnostic about ESG as an alpha source, you should be aware that:

• The EU's Sustainable Finance Disclosure Regulation (SFDR) requires fund managers to classify products by their sustainability characteristics
• Greenwashing (making misleading claims about ESG integration) is increasingly subject to regulatory scrutiny
• ESG data quality remains a challenge, as we noted in the alternative data chapter: different ESG rating providers often disagree substantially on the same company