The Pharmaceutical System
How the modern drug industry actually works. Pricing, research economics, GLP-1 societal effects, and the structural forces shaping what medicines exist and who can afford them.
(IQVIA estimates; the US accounts for roughly 45% of global spend on a much smaller share of population)
(highly contested figure; independent estimates range from a few hundred million to several billion)
(Trilliant Health; figure rising rapidly with extended indications and supply expansion)
A note on framing. The pharmaceutical system is one of those domains where neither the "Big Pharma is uniquely evil" framing nor the "Big Pharma is the heroic developer of life-saving drugs" framing captures what is actually happening. The honest picture has both heroic discoveries and predatory pricing, both genuine R&D risk-taking and aggressive patent-extension manoeuvres, both genuinely curative new therapies and a wellness-industry edge that has crossed over into outright misinformation. This page tries to walk through the structural picture so a reader can place specific stories - a price spike, a new approval, an enforcement action - in context.
How drug development actually works
Modern drug development moves through four roughly defined stages. Discovery, where chemists or biologists identify candidate molecules - traditionally through screening libraries of compounds, increasingly through computational and AI-assisted design. Preclinical development, where candidates are tested in cells and animals for both efficacy and toxicity. Clinical trials, which run in three phases (small safety trial, medium-sized efficacy trial, large-scale confirmatory trial) and consume the majority of the time and cost of the process. Regulatory approval, where the FDA, EMA, or equivalent national regulator reviews the trial data and decides whether to approve. The whole process averages about 10-15 years from candidate identification to approval, with a high attrition rate at every stage.
What this produces is an industry with an unusual cost structure. Most of the spending is on candidates that never reach approval. Most of the candidates that do reach approval take many years to recover their development costs. A small number of approved drugs - blockbusters earning more than $1 billion annually - generate the bulk of profits and effectively cross-subsidise the larger number of more modest products. This concentration is structural; it shapes which diseases get studied, which patient populations get attention, and how aggressively prices are set on successful drugs.
The cost of developing a new drug is one of the most-debated numbers in health economics. The Tufts Center for the Study of Drug Development, which has industry funding, periodically estimates an average of about $2.6 billion per approved drug (with substantial methodology choices). Independent academic estimates - notably by Olivier Wouters and colleagues at LSE - have produced lower figures around $1.3 billion median. Critics including Aaron Kesselheim at Harvard and Marcia Angell argue that the headline figures inflate genuine private-sector spending by including substantial public funding (NIH-funded basic research, NSF grants), capital costs over long horizons, and the implicit cost of drugs that fail. The available evidence indicates that drug development is genuinely expensive and risky, that the headline cost figures are higher than the cleanly-defensible estimates, and that meaningful public-sector contribution exists alongside the private-sector spending the industry foregrounds.
The pricing puzzle
US prescription drug prices are roughly two to four times higher than prices for the same drugs in other developed economies. This is the single most-discussed feature of the modern pharmaceutical system, and it has a relatively well-understood structural explanation: most other developed countries negotiate drug prices through their single-payer or heavily-regulated health systems, while the US has historically prohibited Medicare from negotiating most drug prices and has fragmented private negotiation across many payers with varying leverage.
The Inflation Reduction Act of 2022 began allowing Medicare to negotiate prices on a small subset of drugs, with the first negotiated prices taking effect in 2026 for ten drugs and expanding in subsequent years. The list of drugs subject to negotiation is small relative to the total market, but the precedent is significant. Industry response has included substantial lobbying for narrowing implementation, several legal challenges, and some explicit threats to reduce US R&D investment in response. Whether the negotiations meaningfully reduce US prices over the long term, or are slowly eroded by industry counter-action, is still being determined.
Pharmacy benefit managers (PBMs) - companies that act as middlemen between insurers and drug manufacturers - are a substantial and under-covered part of the US pricing system. The three largest (CVS Caremark, Express Scripts, OptumRx) control roughly 80 per cent of the prescription-drug claims market. Their economic model involves opaque rebates and fee structures that move money in ways that are hard to audit; FTC investigations since 2022 have repeatedly raised concerns about PBM business practices contributing to elevated prices. Reform of PBM operations is one of the rare pharmaceutical-policy areas with bipartisan interest in the US, though concrete change has been slow.
The international pricing question is also more complicated than the headline "Americans subsidise the world" framing suggests. Most countries with negotiated prices pay enough to keep drugs available; pharmaceutical companies still launch in those markets and earn meaningful profit there. The US premium reflects the absence of countervailing buyer power as much as a structural cross-subsidy of global R&D. Whether the premium is "fair" depends on values about how the global cost of pharmaceutical innovation should be distributed, which is a political question more than an economic one.
The patent system and what it actually rewards
Patents are the legal mechanism that lets pharmaceutical companies recover R&D investment by selling at prices well above marginal manufacturing cost. A new chemical entity typically receives 20 years of patent protection from the date of filing, though the practical period of market exclusivity is much shorter - often around 8-12 years from approval - because filing happens early in development and the clock runs through the long clinical-trial period.
What the patent system actually rewards is more specific than "discovery." Companies receive patents on the molecule itself, on specific formulations (sustained-release, combination products), on methods of use, and increasingly on minor variations of approved drugs. The practice of "evergreening" - extending effective patent life through reformulations, combination products, or new indications shortly before the original patent expires - is well-documented. The most cited example is AbbVie's Humira, where a thicket of secondary patents extended US market exclusivity well beyond the original patent expiry; biosimilar competition only began arriving in 2023 despite the underlying biologic being approved in 2002.
Generic competition, when it arrives, typically reduces prices by 80-90 per cent over a few years. The contrast between branded prices during patent exclusivity and generic prices after expiry is stark, and it is one of the more reliable arguments against the "drug development requires high prices" position - many drugs developed under high prices still produce substantial profits at much lower generic prices because manufacturing cost is low. The available evidence indicates that high prices during exclusivity reflect not just R&D recovery but also what the market will bear.
Biologics - drugs derived from living cells, including most cancer immunotherapies and the GLP-1 weight-loss drugs - are harder to copy than small molecules and have correspondingly longer effective exclusivity. Biosimilar competition exists but typically reduces prices by 30-50 per cent rather than 80-90 per cent. As biologics have become a larger share of the drug market, the patent-cliff dynamic that historically reduced costs is weaker than it was a decade ago.
GLP-1s and the Ozempic moment
The GLP-1 receptor agonists - semaglutide (Ozempic, Wegovy, sold by Novo Nordisk) and tirzepatide (Mounjaro, Zepbound, sold by Eli Lilly) - are the most consequential pharmaceutical development of the mid-2020s. Originally approved for type 2 diabetes, the dramatic weight-loss effects led to rapid expansion of off-label and approved obesity uses, and ongoing trials are showing additional benefits in cardiovascular disease, alcohol use disorder, kidney disease, and several other indications.
What makes GLP-1s structurally significant goes beyond any single condition. They are the first widely-effective pharmaceutical for obesity, which affects roughly 40 per cent of US adults and substantial fractions of populations in most developed economies. They appear to reduce craving behaviours more broadly, with implications for food-industry consumption patterns, alcohol consumption, and possibly addictive behaviours generally. Stock-market analysts who follow consumer staples have downgraded several food and beverage companies based on projections of long-term consumption shifts driven by GLP-1 adoption, with mixed evidence so far on whether those projections are correct.
The supply, pricing, and access situation has been chaotic. The drugs are difficult to manufacture at scale; shortages have been persistent through 2023-2025. List prices are around $1,000-1,500 monthly in the US, with insurance coverage varying widely; out-of-pocket costs of $500-1,200 monthly are common. Compounding pharmacies have produced lower-cost copies of varying quality during shortages; the FDA's enforcement against compounding has been contested. International prices are substantially lower (Denmark, the home country of Novo Nordisk, charges roughly $150-200 monthly).
The longer-term societal implications are still being mapped. If GLP-1 adoption continues at current trajectories - and supply expands to meet demand - the obesity rate in developed economies could decline meaningfully for the first time in decades. The downstream effects on healthcare costs, food industry economics, fitness industry economics, and even cultural attitudes about body composition are large enough that several books, research programmes, and policy initiatives are now devoted to studying them. Calley Means and Casey Means, with their wellness-industry-adjacent critique, have argued that GLP-1s are masking underlying lifestyle problems; this argument has both legitimate substance (the drugs are not a substitute for nutrition and movement) and an overstated dimension (the lifestyle-only intervention model has demonstrably failed for most people over decades). The honest position holds both pieces.
The opioid catastrophe and what it documented
The US opioid epidemic, which has killed approximately 800,000 people from 1999 to the present, was substantially driven by pharmaceutical-industry actions documented through extensive litigation and journalism. Purdue Pharma's marketing of OxyContin, beginning in 1996, included claims about low addiction risk that the company internally knew to be false. The Sackler family, which owned Purdue, took roughly $11 billion out of the company in the years before its 2019 bankruptcy. The eventual settlements distributed money to states for treatment programmes but allowed Sackler family members to retain substantial wealth, an outcome that critics including Patrick Radden Keefe (whose Empire of Pain documented the case) argued failed basic justice tests.
The opioid case is significant for the broader pharmaceutical-system understanding because it documented in court how the industry can operate when oversight is weak. The role of pharmaceutical sales representatives in directing physician prescribing, the role of pharmacy benefit managers and distributors in maintaining supply, the role of medical journals in publishing industry-favourable studies, the role of patient-advocacy organisations partly funded by industry in shaping regulatory pressure - all were exposed in litigation. The reforms that followed (prescription drug monitoring programmes, restrictions on direct-to-physician marketing, increased FDA scrutiny of opioid approvals) reduced new prescription rates but did not stop the deaths, which shifted from prescription opioids to heroin and then to fentanyl.
The fentanyl phase of the crisis is largely no longer a pharmaceutical-industry story; the supply is mostly Mexican-cartel-produced from Chinese precursors, covered in the Illicit Economies page on this site. But the pharmaceutical-industry origins of the larger arc remain a significant and recent example of what the system can do when the regulatory and accountability mechanisms fail. The $50+ billion in eventual settlements is genuine accountability of a kind; it is also widely viewed as inadequate to the scale of harm caused.
The broader public-health interface
Pharmaceutical companies are not the only actors that determine what medicines exist and who can access them. The FDA and equivalent international regulators set approval standards. National health systems, insurers, and PBMs determine what is covered and at what price. Academic researchers and the NIH fund a substantial fraction of the basic science that pharmaceutical R&D builds on. Patient-advocacy organisations shape regulatory and political pressure. Journals and the broader medical-publishing system determine what evidence is widely seen.
The relationships between these actors have produced documented patterns of concern. Industry funding of physician education has shaped prescribing patterns in ways that internal-document discovery in litigation has revealed. Industry-funded clinical trials have a documented pattern of producing more favourable results than independent trials of the same compounds. Patient-advocacy organisations - some entirely industry-independent, some substantially industry-funded - have a mixed record on representing actual patient interests versus industry priorities. The broader academic-medical-publishing system has been the subject of ongoing reform efforts (registered trials, open-data requirements) with partial success.
The available evidence indicates that the pharmaceutical system involves substantial conflicts of interest, that the conflicts are partly necessary (drug development is expensive and benefits from industry-academic collaboration), and that the question of how to manage the conflicts is a real ongoing policy challenge rather than a solved problem. Aaron Kesselheim's Program on Regulation, Therapeutics and Law at Harvard is one of the more careful research programmes that documents the system as it actually works rather than as either critics or defenders frame it.
The paths from here
Continued slow reform with industry pushback
Medicare price negotiation expands to a larger drug list. PBM reform proceeds incrementally. Patent-evergreening rules tighten modestly. The industry adapts business models without losing fundamental profitability. The US-international price gap narrows somewhat but persists. This is the most likely trajectory under current political conditions.
GLP-1s reshape the obesity-and-cardiometabolic landscape
Supply expansion catches up with demand. Insurance coverage broadens. Adoption reaches 25-40% of US adults. Obesity rates decline meaningfully for the first time in decades. Healthcare costs decline as cardiovascular events and diabetes complications fall. Food industry profits from highly-processed foods decline noticeably. This is the optimistic GLP-1 case; whether it materialises depends on supply, pricing, and adherence over a long horizon.
A major safety event reshapes regulation
A previously-unrecognised side effect of a widely-used drug class - GLP-1s, recent biologics, AI-assisted-discovery products - emerges at scale. The FDA approval process is reformed in response. Industry trial-design and pharmacovigilance practices change. Medical and political trust takes another hit. The Vioxx case (2004) is the recent precedent; the next case has not been identified but the structural conditions for one persist.
AI-assisted drug discovery accelerates the pipeline
AI tools for protein-structure prediction, molecule design, and trial-design substantially compress the timeline and cost of drug development. New therapies for previously hard-to-target diseases (Alzheimer's, ALS, rare cancers) arrive at higher rates. The industry economic structure shifts - venture capital and biotech startups gain ground; large pharma's role becomes more about distribution and trials than discovery. Whether this translates to lower prices for patients depends on regulatory and competitive dynamics.
Geopolitical fragmentation hits the supply chain
US-China tension extends to pharmaceutical supply. China currently produces a substantial share of active pharmaceutical ingredients globally. Tariffs, export controls, or shipping disruption produces drug shortages. Reshoring efforts (already underway since COVID) accelerate but produce limited capacity in the medium term. Drug prices rise; specific shortages persist.
A wellness-skeptic-political alliance forces deeper change
The combination of populist health-skepticism and progressive critiques of pharmaceutical industry produces unusual political coalitions. Reform that has been politically blocked - Medicare-for-all, broader price negotiation, restrictions on direct-to-consumer advertising (currently allowed only in the US and New Zealand) - becomes possible. The pharmaceutical-industry political position weakens substantially. This is currently a minority outcome but is more plausible than at any point since the early 2010s.
Where serious analysts disagree
The system produces genuine medical progress at high but defensible cost
The pace of pharmaceutical innovation - particularly in oncology, cardiometabolic disease, and rare diseases - has been substantial. The cost is real but reflects genuine R&D risk. Reforms should target specific abuses (PBM opacity, evergreening, direct-to-consumer advertising) without disrupting the basic incentive structure that produces the new drugs.
Held by: the mainstream of the pharmaceutical-industry trade groups (PhRMA, BIO), much of the academic-medical research community, and several health-economists including Joseph DiMasi (Tufts). The case has empirical support in the rate of new approvals; weaknesses in the magnitude of cost differentials.
The system is structurally extractive and needs major reform
The genuine medical progress comes substantially from publicly-funded basic research, with the private sector capturing disproportionate value through pricing power that does not reflect actual development cost. Major reform - Medicare-style negotiation across the board, public R&D pipelines for high-priority drugs, restrictions on patent evergreening - would produce comparable innovation at much lower cost.
Held by: Marcia Angell (former NEJM editor, "The Truth About the Drug Companies"), Dean Baker (CEPR), Mariana Mazzucato ("The Entrepreneurial State"), parts of the academic health-policy community. Has empirical support in the public-funding-of-basic-research data and weaknesses in fully accounting for genuine private-sector R&D risk.
The ultra-processed food industry is a larger health driver than pharmaceutical access
The chronic-disease epidemic that pharmaceutical drugs partly treat is largely a consequence of the modern food environment. Reforming pharmaceutical access without reforming the food system addresses symptoms rather than causes. GLP-1s are dramatic but expensive interventions that compensate for an environment that produces obesity. The right framing is food-and-pharmaceutical-system reform together, not separately.
Held by: Calley Means and Casey Means (whose framing has reached substantial political prominence), parts of the public-health-prevention community, Robert Lustig (UCSF), Mark Hyman. Has empirical support in the chronic-disease rate trajectory and weaknesses in cleanly attributing causation to specific food-environment changes.
PBMs are the most consequential underregulated actor
The opacity of pharmacy benefit manager pricing and rebate structures is responsible for a substantial fraction of US drug-pricing dysfunction. The "Big Three" PBMs (CVS Caremark, Express Scripts, OptumRx) extract value from both manufacturers and pharmacies in ways that are hard to audit. Meaningful reform of PBM operations would have larger effects on US prices than any other single intervention.
Held by: the FTC under recent leadership, several state attorneys general including those in California and Texas, parts of the bipartisan congressional coalition pushing PBM reform legislation. The case has empirical support in FTC investigations and limitations in the technical complexity of PBM operations.
Conflict-of-interest concerns are systemically underweighted
Industry-funded education, industry-influenced clinical trials, industry-supported patient advocacy, and industry-funded medical journals collectively produce a research and prescribing environment that is more shaped by industry interests than the medical profession acknowledges. Reform requires structural separation of industry funding from physician education, prescribing decisions, and research evaluation.
Held by: Aaron Kesselheim (Harvard), the JAMA editorial position under recent leadership, Adriane Fugh-Berman, the broader BMJ and Cochrane evidence-quality community. Has empirical support in study-by-funder effect-size analyses; weaknesses in identifying alternative funding structures that would produce equivalent research.
None of these readings is fully right or wrong. What can be said from the available evidence: the pharmaceutical system produces genuine medical progress alongside genuine pricing dysfunction; the cost-of-development numbers are real but inflate easily under industry framing; PBMs are a substantial and under-regulated source of US-specific dysfunction; conflict-of-interest concerns are real and structurally hard to fix; GLP-1s represent a possibly-transformative shift whose downstream effects are still being mapped; and reform that has been politically blocked for decades has more openings now than at most points in recent history, though the durability of any specific reform remains uncertain.
What this means for you
If you take prescription medications
Generic substitution, when available, typically reduces cost by 80-90 per cent without therapeutic difference. For drugs without generic alternatives, manufacturer patient-assistance programmes are real and underused; pharmacy-pricing-comparison tools (GoodRx, Cost Plus Drugs) often produce substantially lower out-of-pocket prices than insurance-mediated pricing for many drugs. Asking your pharmacist directly about lowest-cost options is reliably useful and is often not the default the pharmacy system offers.
If you are considering GLP-1 use
The decision is genuinely consequential and deserves substantive medical evaluation, not just internet research. The drugs work for most people who tolerate them; the side effects are real and meaningful; long-term durability of weight loss after discontinuation is poor without behavioural changes; cost and supply remain volatile. Your physician's individual experience with these drugs varies; consulting with a primary-care physician who has substantial GLP-1 experience produces better outcomes than relying on telehealth services that have less continuity.
If you encounter wellness-industry health advice
The boundary between legitimate critique of the pharmaceutical system and outright misinformation has been blurred deliberately by some major wellness-industry actors. The legitimate critique - that food environment matters, that lifestyle interventions are under-emphasised, that pharmaceutical solutions can mask underlying problems - is genuine. The misinformation - that vaccines cause autism, that nearly all medical advice is corrupt, that you can solve cancer with diet alone - is dangerous. Distinguishing them requires more discernment than mainstream coverage usually conveys; sources that engage critically with both pharmaceutical-industry capture AND wellness-industry misinformation (Eric Topol's coverage, the Institute for Functional Medicine's careful end) are more reliable than either pure-defender or pure-critic positions.
If you read pharmaceutical-industry coverage
STAT News, Endpoints News, and Bloomberg's pharmaceutical coverage are the most reliable specialised sources. Investor-facing sources (Bloomberg, Financial Times) tend to cover the industry's economic and strategic moves more honestly than consumer-facing sources, which often present industry framings without sufficient critical lens. Academic medical journals (NEJM, JAMA, BMJ) are reliable on therapeutic evidence but less so on industry economics. Reading across registers produces better understanding than any single source.
If you vote on pharmaceutical-related policy
The most consequential current policy areas are Medicare price negotiation expansion, PBM operational reform, patent-evergreening restrictions, and direct-to-consumer advertising rules. These are unglamorous and consequential; engaging them substantively requires reading legislation rather than headlines. The political-economy reality is that pharmaceutical industry political contributions are large and bipartisan, which means that nominal political affiliation predicts position on these questions less reliably than other policy domains.
