Picture the moment. A trade association's chairman stands at a lectern, assures a roomful of journalists that the industry takes conduct seriously, and announces a new voluntary code. The journalists write it down. The members file out. Nothing changes.
Some industries avoid that scene entirely. They build bodies with actual teeth, maintain standards that cost something to meet, and expel members who cheat. Others run the voluntary-code pantomime on a loop until a regulator finally loses patience. The difference isn't virtue. It comes down to three structural conditions: how visible the harm is, how concentrated the players are, and whether the industry stands to lose more from a scandal than from the price of compliance. Get those three conditions wrong and no trade association on earth will hold the line.
When the club actually works
Take the accounting profession. For most of its modern history, the major chartered accountancy bodies in Britain and their equivalents elsewhere maintained genuine disciplinary authority. Members who cooked books got struck off. Standards were published, examinations were standardised, and a firm expelled from the institute couldn't credibly pitch for audit work. That worked because the pool of credentialed practitioners was relatively small, the reputational harm of misconduct was immediate and traceable, and the clients buying the service cared deeply about the credential itself. The badge meant something because the body that issued it had teeth.
Contrast that with the financial products industry in the decades before statutory consumer protection frameworks took hold. The players were numerous, the products were opaque, the harm to individual buyers was diffuse and slow to surface, and the profits from misbehaviour were enormous. A trade association issuing stern guidance was, in practice, a polite letter nobody had to open.
The conditions for self-regulation simply weren't present.
The mechanism matters here. Self-regulatory bodies run on the logic of repeated interaction: the same firms meet each other at the same table, year after year, and the cost of cheating accumulates over time. Game theorists call this the shadow of the future. When the shadow is long, cooperation is rational. When it's short, because the market is fragmented, entry is easy, or a single transaction pays enough to absorb any reputational damage, defection wins every time. Payday lending is the textbook case. Dozens of small operators, customers who are price-sensitive and often desperate, a harm that takes months to fully materialise. Studies tracking arrears in that market found the median borrower didn't recognise the full cost of rollover until three or four cycles in, by which point the original lender had already collected its margin and moved on. No trade body was ever going to fix that. Statutory caps on interest rates and mandatory affordability checks were the only lever that moved behaviour.
What people almost always get wrong
The common assumption is that self-regulation fails because industries are corrupt or cynical. Sometimes that's true. But the more common failure is structural, not moral, and the distinction matters enormously for how you design a response.
Consider two solicitors. One works in a regional town; one sits at a large city firm. Both belong to the same professional body. The city solicitor handles complex commercial work, and a disciplinary finding would cost her millions in lost mandates. The regional solicitor runs a high-volume conveyancing practice. His clients don't read the register of disciplinary findings and wouldn't know where to look. The same rulebook, working with unequal materials, produces entirely different incentive structures. The body isn't corrupt. It's just applying a single set of rules to a profession that isn't actually a single thing.
The pharmaceutical industry offers the inverse lesson. Drug manufacturers operate under some of the most demanding self-imposed testing and disclosure standards of any sector, not because they are uniquely virtuous, but because the commercial logic of clinical trial integrity is overwhelming. A falsified trial that reaches market and harms patients doesn't just destroy a product. It can erase a company's entire pipeline, trigger criminal liability for named executives, and wipe ten figures off the market capitalisation in a single trading session. The harm is visible, traceable, and career-ending. That is the environment in which voluntary standards flourish: not goodwill, but aligned consequences.
You can almost predict which industries will eventually need statutory intervention by asking one question: if a firm cheats, how long before anyone finds out, and who bears the cost? When the answer is "years, and mostly the customer," self-regulation is furniture. When the answer is "immediately, and mostly the firm," a trade body can do real work.
Ask yourself: how many industries currently lobbying against new rules would actually pass that test?
Statutory intervention isn't a sign that an industry is uniquely bad. It's a sign that the structural conditions for self-policing were never present to begin with. The interesting question, and the one regulators consistently underinvest in asking, is whether those conditions can be engineered into existence: through mandatory disclosure, through liability rules that shorten the distance between misconduct and consequence, through market design that makes cheating expensive rather than merely frowned upon. Sometimes the answer is yes, and when it is, a lighter regulatory hand becomes genuinely possible. The prize is real. But the starting point is an honest audit of the structure, not a faith in the trade association's letterhead. Get that audit wrong and you are not deregulating. You are just waiting.