3) Principal–Agent (PA) theory in drug markets — an expanded essay

Back to main article

Core idea. Principal–agent problems arise when a principal (regulator, owner, customer) cannot perfectly observe an agent’s effort or quality choice (bartender, retailer, street seller). Hidden action creates moral hazard; hidden information creates adverse selection. Incentive contracts, monitoring, and reputation are the classic tools—yet they work very differently in licit versus prohibited markets. (Holmström 1979; Laffont & Martimort 2002.) (jstor.org)

---

3.1 Legal alcohol (regulated supply chain)

Who are the principals and agents?

• Regulators ↔ producers/retailers (compliance with pricing, age checks).
• Firms ↔ staff (bartenders follow ID rules vs. maximize speed/upsell).
• Customers ↔ bartenders (quality, safety, and not being overserved).

The hidden actions that matter.

• Overservice (serving obviously intoxicated patrons).
• Weak ID checks (serving minors or those without proof).
• Promotion design (deep discounts, “two-for-one” shots) that externalize harm.

Why the problem persists. In multitask settings, paying strongly for sales speed can crowd out attention to hard-to-measure compliance. PA theory predicts that when one task (sales) is easy to measure and another (careful ID/overservice judgment) is not, high-powered incentives push effort toward the measurable task—unless contracts and monitoring are redesigned. (Holmström & Milgrom 1991.) (OUP Academic)

Instruments that work in legal markets (because courts enforce them).

• Excise / Minimum Unit Pricing (MUP): remove ultra-cheap, high-ABV formats that tempt overservice and intoxication spirals at the margin.
• Overservice liability (dram shop–type rules): shifts expected costs back to firms; aligns firm incentives with public safety.
• Random compliance checks and meaningful fines: raise the expected cost of shirking on ID rules.
• Server training + job design: attach explicit evaluation (and rewards) to compliance, not just throughput—classic PA “multitask” fix. (General PA mechanisms per Holmström 1979; Holmström & Milgrom 1991.) (jstor.org)

---

3.2 Illegal heroin (prohibited supply chain)

Who are the principals and agents?

• Importers/wholesalers ↔ street sellers;
• Sellers ↔ customers.

The hidden actions that drive harm.

• Adulteration (unobservable purity), skimming, misreporting, risky sales locations.
  Customers cannot verify quality ex ante; sellers cannot rely on courts to punish cheating. This is textbook adverse selection—the “lemons” logic—pushing average quality down and variance up. (Akerlof 1970.) (OUP Academic)

How governance shifts without courts.

• Enforcement of “contracts” relies on reputation, surveillance, and violence; organizations remain small, ephemeral, and disorganized because suppressing competition is costly and risky. The result is volatile prices and purity, with risk passed to users. (Reuter 1983; MacCoun & Reuter 2001.) (MIT Press)

Partially legalizing the interface reduces PA failures.

• Medication-assisted treatment (OAT) moves the crucial buyer–supplier interface into licit medicine, collapsing quality uncertainty for a share of demand.
• Drug checking (on-site or at services) introduces third-party verification and early warning—an institutionalized reputation/quality device.
• Supervised consumption sites add observability for the riskiest moments and connect people to care. Evidence and public-health reviews in Europe and North America describe these as harm-reduction infrastructures; findings on population-level mortality vary by setting, but the mechanism—reducing information and execution risk at the point of use—is squarely PA-consistent. (EMCDDA/EUDA on drug checking; Lancet Toronto spatial analysis; Canada 2016–2024 review.) (euda.europa.eu)

---

What PA theory changes in practice (side-by-side takeaway)

Feature	Alcohol (legal)	Heroin (illegal)
Enforcement	Courts, inspectors, fines	Reputation, surveillance, violence
Hidden action	Overservice; weak ID	Adulteration; skimming; risky spots
Monitoring	Mystery shoppers; audits	Spot tests; word-of-mouth; turf control
Contracting	Feasible (licenses, liability)	Infeasible → informal coercion
Policy levers	MUP, liability, compliance metrics for staff	OAT, drug checking, supervised consumption to legalize interfaces
Expected effect	Shift firm/worker incentives toward safety	Collapse quality uncertainty; reduce execution risk

Bottom line. In licit markets, PA tools are contracts, audits, and liability. In prohibited markets, PA tools become reputation and force—which are blunt and dangerous. Building legal interfaces (OAT, checking, supervised use) installs institutional monitors that shrink information asymmetries and moral hazard where harm is highest. (Laffont & Martimort 2002; MacCoun & Reuter 2001.) (De Gruyter Brill)

---

Literature & Links (APA)

• Akerlof, G. A. (1970). The market for “lemons”: Quality uncertainty and the market mechanism. Quarterly Journal of Economics, 84(3), 488–500. Oxford Academic · JSTOR. (OUP Academic)
• Holmström, B. (1979). Moral hazard and observability. Bell Journal of Economics, 10(1), 74–91. JSTOR · author/offprint. (jstor.org)
• Holmström, B., & Milgrom, P. (1991). Multitask principal–agent analyses: Incentive contracts, asset ownership, and job design. Journal of Law, Economics, & Organization, 7(special issue), 24–52. Oxford Academic · PDF. (OUP Academic)
• Laffont, J.-J., & Martimort, D. (2002). The Theory of Incentives: The Principal–Agent Model. Princeton University Press. Publisher extract (PUP via De Gruyter Brill). (De Gruyter Brill)
• MacCoun, R. J., & Reuter, P. (2001). Drug War Heresies: Learning from Other Vices, Times, and Places. Cambridge University Press. Publisher page. (Cambridge University Press & Assessment)
• Reuter, P. (1983). Disorganized Crime: The Economics of the Visible Hand. MIT Press. Publisher page. (MIT Press)
• European Union Drugs Agency (EUDA/EMCDDA). (2023). Health risk communication strategies for drug checking services. Spotlight · Report (PDF). (euda.europa.eu)
• Russell, C., et al. (2023). Overdose mortality incidence and supervised consumption services in Toronto, 2017–2019. The Lancet Public Health, 8(12), e1086–e1095. Publisher page. (The Lancet)
• Public Health Agency of Canada. (2025). Supervised consumption sites and population-level overdose mortality: Systematic review (2016–2024). Government report (PDF). (canada.ca)

---

Publishable version of the prompt

“Please expand ‘3) Principal–Agent (PA) theory in drug markets’ into an essay that contrasts legal alcohol and illegal heroin supply chains. Identify principals/agents, hidden actions, and instruments; explain why multitask incentives matter for overservice/ID checks; and show how OAT, drug checking, and supervised consumption partially legalize the interface in prohibited markets. Provide APA references with publisher-first links.”

Prüfprotokoll

• Status: Enriched section v1.0 (WordPress-ready).
• Checks: Core PA claims grounded in Holmström (1979), Holmström & Milgrom (1991), Laffont & Martimort (2002); market-specific claims referenced to Reuter (1983) and MacCoun & Reuter (2001); interface interventions linked to EUDA, Lancet, and PHAC.

Back to main article