Written: March 8, 2018 | Released: July 2, 2021
Guilds are common and enormously influential today. We’re so used to the way that society is organized that it’s easy not to notice what a “guild-based” society we have. While guilds provide major benefits, they also come with societal costs that I think are almost always under-appreciated.
What I’m referring to here are professional groups that have achieved a monopoly or near-monopoly on providing a certain type of service. (Note: I’ll be focusing here on the United States in particular, and a lot of details will be different in other countries, though I would expect that many of the principles of guilds are the same across countries.)
Understanding these guild systems can help us answer questions about how the U.S. is organized that at first seem quite baffling, like these:
• In law: why is it so expensive to hire someone? Average hourly rates for associate lawyers in the U.S. at large firms appear to range from about $150 per hour (e.g., insurance) to $360 per hour (e.g., mergers and acquisitions) (see here for 2015 numbers). Compare this to the average hourly wage in the U.S., which is closer to $20 or so per hour (source). Since the only realistic way to use the court systems (and also to understand the laws in some industries) is to hire lawyers, the extremely high cost of lawyers puts a tax on society. Recall that prices for labor are not based on the intrinsic difficulty of the work per se; they are based on the balance of supply and demand. So why then is the supply of lawyers (that people want to) hire so small?
• In medicine: why is it that after undergoing four years of grueling medical school training at great personal expense (e.g., $200,000 – $300,000), medical residents in the U.S. then work painfully long hours for 3-7 additional years at an absurdly low hourly pay? Residents appear to work about 80 hours per week on average, with shifts sometimes lasting 30 hours straight and with about two-thirds of first and second-year residents reporting that they sleep fewer than six hours per night (source). Some sources claim that medical residents work longer than 80-hour weeks, which would be in violation of regulations. Wikipedia says, “…in many locations, trainee doctors commonly work 80 to 100 hours a week, with residents occasionally logging 136 (out of 168) hours in a week.” So why is residency such a brutal undertaking?
• In academia: why are there vastly more slots available to get a PhD in the U.S. than there are academic jobs available for PhD graduates in fields where a PhD prepares you for few positions other than an academic job? The median time from the beginning of graduate education until getting a PhD in the U.S. is very long (six years for physical science, seven years for math, nine years for humanities, 12 years for education), yet the academic job market is generally known to be brutally competitive in most disciplines. The situation appears to be particularly bad in non-science/engineering fields, where apparently only about 12% of doctorate recipients have a U.S. based postdoc lined up when their PhD is awarded, with about 55% of those in the humanities/arts having any kind of job lined up at all (as of 2015). Some PhD graduates will end up getting postdocs later, while others end up finding jobs as massively overqualified teachers (see the source of these statistics here). So why are so many PhDs created in fields where job prospects are extremely bleak?
• In many fields requiring licensing or a special degree: why do the required training programs typically teach so much material that will not be used once the students actually enter that field? It’s tough to get hard data on how prevalent this problem is, but when informally talking to lawyers, doctors, and academics, it seems pretty clear that the significant majority of their training ends up not being actually used by them once they enter their chosen field. So why then isn’t the training cut in half with faster specialization?
Here’s my attempt to break down how these modern guilds work, the advantages they bring, and the problems they create.
Caveat: this is a complex topic, I am not a lawyer, doctor, or academic (though I do have a PhD in math), and there is still plenty about this that I don’t understand. What follows is my current best attempt to make sense of these “guild” systems.
BACKGROUND
In medieval times, guilds were associations of merchants or craftspeople, which maintained standards of quality while protecting their members’ interests (for instance, by influencing laws related to the services offered and creating a monopoly on production).
In modern times, what I’m calling “guilds” are the associations of professionals or workers, which maintain standards of quality while protecting their members’ interests (for instance, by influencing laws related to the services offered and creating a monopoly on production).
Modern and medieval guilds, of course, differ in a number of ways, given how much society has changed, but I think the parallels are close enough that the term “guild” is worth applying to modern cases.
FORMATION
Here’s one explanation for why guilds (whether ancient or modern) might form in the first place. Suppose there is a proliferation of low-quality offerings for a particular service (e.g., self-proclaimed “surgeons” who make up their own dangerous yet useless surgical procedures, which they then persuade injured people to undergo). Consumers are unhappy because they can’t easily tell which services are helpful or harmful.
Those who are above average at providing the service share the incentive with consumers of filtering out the low-quality providers, as doing so will reduce competition and supply while raising salaries and protecting the prestige of their field.
A quality filtering mechanism might also be appealing to lawmakers since they may want to please consumers; plus, if the field sets standards for quality itself, then lawmakers don’t have to take on that complex task.
Finally, providers would generally prefer to regulate themselves than have lawmakers force rules onto them. If providers can self-regulate to appease lawmakers, it enables that budding guild to choose rules that promote quality and societal welfare while also benefiting the guild members themselves.
Hence the formation of a guild initially may solve significant problems simultaneously for consumers, high-quality providers, and lawmakers.
In one case, one can imagine a guild being spurred into creation by regulators cracking down on harmful practices. In another, one can imagine a guild getting started as a reaction to consumer backlash. And one can also imagine a guild being formed by a conscious effort on the part of the high-quality offerors of service to block the lower-quality competition.
But whatever the initial cause, as we’ll see, what may start as the solution to a problem can eventually grow into a problem itself.
GUILDS VS. UNIONS
Guilds can easily be confused with unions, and while they share some commonalities, they typically differ in a number of ways. What they have in common is that they are both formed when people who engage in similar work join together to better look out for their own interests.
But unions are typically groups that join together to gain power relative to the large companies that employ them, whereas guilds typically join together to gain power relative to their competitors (e.g., those offering low-quality or cheap variants of their service).
Some other typical differences are that:
• guilds are more likely to be for very highly-skilled jobs
• guilds typically keep their membership size constrained on purpose (to incur greater benefits to existing guild members), whereas unions are more likely to try to grow their membership (to increase negotiating power)
• modern guilds are more likely to be national in scope, whereas modern unions are more likely to have large amounts of variation by region
• guilds are more likely to be entrenched via regulation and receive government funding, whereas unions are more likely to have only generic protection (as a class) if they have protection at all
• guilds seem to have largely maintained their prominence, whereas unions seem (at least in the U.S.) to have been diminishing in power
THE BENEFIT OF GUILDS
The main societal benefit from guilds is quality control, which comes in a variety of forms.
Guilds may:
• reduce the number of fake products or services (e.g., by having production by non-guild members completely banned, or by setting quality standards for production)
• standardize the length and nature of training (e.g., by the creation of schools and exams)
• provide extra options for retaliation for those who are cheated by a guild member (e.g., by creating procedures for challenging a person’s fitness to be a guild member and by establishing standards of ethics)
• define best practices for their field
• work with or influence regulators in the creation of useful laws governing behavior in their field
THE GUILDS OF TODAY
Perhaps the three most powerful and prestigious guilds we see today in the U.S. are those of lawyers, doctors, and academics. There are a ton of other guilds in our society as well, including dentists, pharmacists, clinical psychologists, social workers, accountants, investment advisors, barbers, cosmetologists, etc. Usually, a form of work that requires a license or degree requirement in order to be legally allowed to practice it is a form of guild, though guilds differ dramatically in how powerful and how unified they are.
It’s important to note that most members of guilds do not directly serve the interests of the guild, and guild members don’t necessarily even identify as being a part of it (or even recognize that a “guild” exists).
Generally speaking, in joining the guild, they will have entered a system established long ago that they had no part whatsoever in creating. Furthermore, most guild members are never in a position of power that allows them to make decisions on behalf of the other members.
Typically, there is one (or just a few) governing bodies that act on behalf of the members’ interests. The most important power of these governing bodies is that they define unambiguous criteria for guild membership. You are either a member or not, as determined by a set of criteria they invent, usually involving a standardized form of training followed by a sufficiently high score on one or more exams. Typically, though, it will be a very small number of past and current guild members that are responsible for the workings and rules of the guild.
WHAT GUILDS GIVE THEIR MEMBERS
Usually, guild membership involves increased incomes (by keeping guild membership small and keeping out competitors through licensing or degree requirements). But guilds might instead provide job security (e.g., tenured professors are almost never fired and have great freedom to work on whatever interests them) and increased prestige. Prestige partly comes from keeping membership small, but also from using very challenging entrance exams (so that people believe guild membership is linked to high intelligence), kicking sketchy or scammy practitioners out of the guild (so they don’t tarnish the reputation of the guild), etc. Sometimes guilds also provide low-cost, highly-skilled labor to their members in the form of “apprentices” who are trying to get into the guild (e.g., think PhD students).
Of course, it’s not like there is some scheming, secret committee that plans out what to give guild members at the expense of non-members. It’s that a certain set of incentives (aligned among certain providers of a service) tends to produce certain effects (e.g., the choice to restrict membership, and to block others from providing the service when possible, which creates artificially increased prices for the service and increased prestige, and so on).
GUILDS WITHIN GUILDS WITHIN GUILDS
Large guilds sometimes subdivide into smaller guilds that have even more specific interests. For instance, doctors are a guild, but orthopedists, a subset of doctors, might reasonably also be thought of as a guild if they sometimes act as a unit with common interests and have their own rules for certification. Academia is a particularly interesting example because it occasionally acts like one large guild, whereas in most instances, it is more like dozens of small guilds (representing different academic disciplines).
Another form of guilds within guilds are sub-guilds based on state. For instance, lawyers and social workers who wish to switch the U.S. state in which they practice will often have to jump through extremely annoying hoops in order to do so (e.g., by retaking a slightly different set of exams).
For instance, in law, guild standards are set by bar associations that are per state and which have a rather complex integration with the government. According to Wikipedia:
“In the United States, admission to the bar is permission granted by a particular court system to a lawyer to practice law in that system. This is to be distinguished from membership in a bar association. In the United States, some states require membership in the state bar association for all attorneys, while others do not. Although bar associations historically existed as unincorporated voluntary associations, nearly all bar associations have since been organized (or reorganized) as corporations. Furthermore, membership in some of them is no longer voluntary, which is why some of them have omitted the word ‘association’; and merely call themselves the ‘state bar’ to indicate that they are the incorporated body that constitutes the entire admitted legal profession of a state…Such an organization is called a mandatory, integrated, or unified bar and is a type of government-granted monopoly.”
Guilds within guilds exist for much the same reason that guilds do (to protect the interests of a narrow group against competition while enforcing quality control).
GUILD BOUNDARIES
While guilds typically have some sort of required license or degree, it’s not necessarily all such license or degree holders that a guild is serving (and guilds certainly aren’t serving those currently in the process of being certified). Take academia, for instance. You pretty much can’t become a tenured professor if you don’t have a PhD, but most people with PhDs aren’t really in the academic guild. I think it’s most natural to think of the academic guild as only including those with tenure and those who are in tenure track positions. Those with PhDs in non-tenure-track teaching positions at universities are often hoping to get into a tenure track position, but I think they derive only very limited benefits from the guild and therefore shouldn’t really be thought of as being “in it.”
Another example is law. If you are technically a lawyer, but you don’t actually practice law (or you’ve let your licensing lapse), then you’re not in the guild in a meaningful way to the degree that you were before (in the sense that your interests are now not as aligned with that of the guild). You may derive some benefits from the guild (such as the prestige of a law degree) but not as many. In some other cases, the prestige and benefit of guilds get very lopsided towards certain types of members (e.g., those that attended only the very most prestigious licensing programs).
PODIATRISTS ARE NOT WHO YOU THINK THEY ARE
Podiatry (a branch of medicine related to disorders of the foot and ankle) is an interesting example of how arbitrarily and rigidly structured guilds can be. A lot of people think that podiatrists are medical doctors (MDs), but actually, in the U.S., they are Doctors of Podiatric Medicine (DPMs) and have a separate guild that is all their own.
There are only nine colleges of Podiatric Medicine, and they must be accredited by the Council on Podiatric Medical Education. To become a podiatrist in the U.S., you have to go to one of these specific colleges for four years of post-grad education, pass a set of specific podiatry exams, and complete at least one year of supervised postgraduate hospital training.
One could easily imagine a world where podiatrists were simply another type of medical doctor, but that’s not the way it shook out. What do you think would happen today if a large number of podiatrists started treating conditions unrelated to the foot or ankle? You can bet they would rapidly get smacked down, either by lawsuits or criminal proceedings, or new regulation. I stumbled on a discussion about podiatry on a med student forum while researching and was struck by various comments expressing condescending vitriol toward podiatry. One commenter even proclaimed that doctors wisely chose not to allow podiatrists into their group because the work is beneath them.
GUILDS EVERYONE PAYS FOR
It’s hard to think of something more beneficial for a group than making it illegal for other people to offer what that group offers. But some guilds have managed to go a step further by getting the government to directly buy their service on behalf of society! The government directly funds the medical guild by paying the salary of residents. It does it in law by paying for court-appointed lawyers. It does it in academia through numerous grants that can pretty much only be won if you’re applying from within a university. Guilds have woven themselves right into society.
Of course, saying that the government pays guilds really means that your tax money pays guilds. This is not necessarily a bad thing (e.g., you might be happy to have some of your tax money go to fund science or to pay for court-appointed lawyers for those who can’t afford their own), but it is not necessarily what you expect your tax money to be doing (e.g., academic grants in areas that you don’t see as valuable, or funding medical residents that you might assume would be paid by the hospitals they work for).
THE SOCIETAL COSTS OF GUILDS
As I mentioned, guilds typically bring the important benefit of quality control, but they also bring with them some significant (and I think often neglected) costs. There is nothing nefarious about these costs: they are typically just the result of guilds following their natural incentives, which just happen to do harm in some cases.
Cost 1: Guilds constrain membership, producing excessively high prices
Guild members have an interest in protecting the guild’s prestige, income, and customers. If guild membership is limited and quality is kept high, guild members earn greater salaries, and guild membership is viewed as more prestigious.
On the other hand, if nearly anyone who wanted to enter a guild were allowed in, then the average guild member would provide lower-quality services (which would be likely to reduce public opinion of guild membership), and wages would fall (due to increased supply of the service without a corresponding increase in demand).
Societally, the problem with keeping guild membership highly restricted is that those who want guild services have to pay high prices for them (e.g., if you hire a lawyer or doctor), and in some cases, may have to wait a long time for service (e.g., if you try to book an appointment with a medical specialist). In the worst case, people may not be able to get service at all (e.g., someone seeking a certain type of medical specialist in a rural area).
So, insofar as guilds have the power to do so, they will want to keep their membership more limited than would be best for society. If we were optimizing purely for societal benefit (but assuming for the moment that the guild has to exist), then we’d want to set guild membership based on the optimal societal tradeoff between quality and quantity (i.e., stop increasing membership when the marginal benefit from increasing quantity matches the marginal cost from decreasing quality). But existing guild members generally have an incentive to keep quantity more constrained than this optimal tradeoff point since they get the same personal benefits as everyone else from increased quantity but incur much higher costs than non-guild members do (e.g., reduced salaries and prestige).
Cost 2: Guilds eat their young
Current guild members do not have aligned incentives with those who wish to join the guild. And since, as we discussed earlier, the number of guild members is generally kept artificially low, there is likely to be an excessive amount of interest in joining the guild relative to the number of open positions.
In practice, interest in joining the most prestigious guilds is so large that they can impose high costs on those who want to join and still have plenty of high-quality applicants. Importantly, the guild would benefit by imposing these costs even if the costs had nothing to do with being a good guild member (since any cost, no matter how arbitrary, will restrict guild membership). But the guild benefits even more if these imposed costs on guild applicants trickle up to guild members (e.g., in the form of application fees or dues or training fees or education costs), or if these “apprentices” pay with their time, serving the guild as a cheap supply of labor.
The practice of apprentices serving existing guild members has gotten extreme in some cases. For instance, as mentioned previously, in some areas of academia, it’s common to spend 5-9 years as a low-paid PhD student doing work that your advisor chooses despite having very weak job prospects in the field upon graduation. Some of this work is surprisingly manual and repetitive in nature (e.g., some types of biological lab work). Some PhD students claim their advisors purposely delay their graduation to squeeze more cheap labor out of them, though presumably, their advisors would say this is false.
Medicine is perhaps even more egregious in this regard than academia, with four years of med school and 3-7 years of residency, much of which involves working extremely long hours at a relatively very low salary (given the number of hours and amount of skill involved). Fortunately, these residents are not doing anything important while massively sleep-deprived, like, say, conducting surgery. Oh wait: that’s exactly what some of them are doing. Since average resident salaries are apparently about $55,000 per year, an 80-hour workweek would imply a pay of just $14 per hour before taxes (remember that this is AFTER you’ve completed four years of medical school training). Brutal.
Costs to become a lawyer are very considerable too. You have to take the LSAT, apply to schools, spend three years in law school (with a cost range of maybe $100,000 to $325,000 depending on the school and how you finance it; see here), and pass the bar exam for your state.
A potentially more twisted incentive is caused by the fact that guilds sometimes benefit from training apprentices (since they can collect fees and work from these apprentices), but they don’t benefit from admitting large numbers of new guild members. Hence, some guilds train far more people than they end up admitting into the guild. Academia is perhaps one of the biggest offenders in this regard, in that the number of available tenure track postdoc jobs is usually much lower than the number of newly minted PhD students, meaning that after years of training and years of working for their advisors, most students will be forced to leave their chosen academic field. This is not necessarily that bad when students are learning a skill with high market demand (e.g., computer science), but in areas with low market demand, it can be highly problematic and stressful for students.
In biomedical sciences, in 2014, there was allegedly only about one tenure-track position in the U.S. for every 6 PhD graduates (source). And some fields have it even worse. As one professor of American literature put it in 2009 (source), “Nearly every humanities field was already desperately competitive, with hundreds of applications from qualified candidates for every tenure-track position. Now the situation is becoming even worse.” More generally, a 2013 analysis (source) suggested that there are only enough permanent academic positions for 13% of PhDs to attain such positions. Of course, thankfully, there are other types of jobs for those with PhDs (for instance, other types of educational jobs or industry jobs). Though apparently, only about 60% of PhD students had a job lined up upon graduation as of 2014 (source) across fields.
PhD students who plan to stay in academia and who are not coming out of the most elite universities are especially screwed. For instance, according to research conducted on academic jobs in business, computer science, and history (source), “just a quarter of all universities account for 71 to 86 percent of all tenure-track faculty in the U.S. and Canada in these three fields. Just 18 elite universities produce half of all computer science professors, 16 schools produce half of all business professors, and eight schools account for half of all history professors.”
Another depressing thing about academia is how graduating PhD students that want to stay in academia have absurdly little control over where they end up living. They frequently end up living in isolation in a random town or city where they don’t know a single person beyond their new work colleagues.
So, you see, guilds tend to eat their young.
Cost 3: Guilds require excessive and irrelevant training
In my experience, if you ask lawyers how helpful law school was for the job they actually do, they’ll say it was only a little helpful. And if you ask how similar the bar exam was to either law school or the job they do now, they’ll again usually say only a bit. Likewise, if you ask doctors how much of the huge amount of material they learned in medical school is useful to them now, in my experience, they’ll say that only a small fraction of it is.
As we’ve discussed, guilds benefit from keeping their membership artificially constrained, and they usually achieve this by requiring extensive training and licensing. Unfortunately, there is not much of an incentive (at the guild level) to make this training the best that it can be. Only doctors can practice medicine, so if you don’t like how they are trained and boycott, tough luck, no medicine for you. And if you enter medical training and already know what specialty you want to pursue, that sucks because you still have to rotate through a bunch of irrelevant other areas of medicine.
From the guild’s point of view, it’s important that training and licensing maintain the prestige of the field and keep membership small, but that doesn’t necessarily imply making the training as useful as possible. For instance, it may be easier to maintain prestige by making the training very intellectually demanding than by making it extremely useful or pragmatic. And since it’s really quite challenging to produce highly useful training, without a strong pressure to make it as useful as it can be, we shouldn’t be too surprised how irrelevant or ineffective much of it is.
It’s not that guild training is useless (it’s far from useless). It’s just not what you would design if your goal was to train people efficiently or to reduce costs.
Cost 4: Guilds block alternative perspectives
Since guilds typically have a monopoly or near-monopoly on providing their service and simultaneously control training for that service, they can produce a state of affairs where everyone in the field has had similar education. This tends to make opinions more uniform and can crowd out alternative/challenging perspectives. What’s more, once training has been standardized, it is likely to be updated only very gradually.
For instance, if you have radical new brilliant ideas in medicine that fly in the face of what’s currently taught, good luck getting those ideas taught in medical schools. And if you try to put your radical ideas into practice and are not a medical doctor, you’ll end up in jail.
Sometimes blocking radical new ideas can be highly beneficial to society (e.g., when these ideas are false and would actually hurt a lot of people, or when the potential harms are not yet well understood), so this filtering of ideas can be a real benefit of guilds too. The problem is, the guild’s training programs are likely to adapt very slowly to even really good new ideas, and the guild’s perspective is likely to be a conservative one of continuing to do what has worked well for the guild for a long time, rather than taking the risk and unpredictability of shaking things up. A monopoly naturally has reasons to be extremely risk-averse.
CONCLUSION
Guilds play a major role in society, despite them being rarely discussed as a type of entity. An understanding of guild incentives can help us explain some otherwise baffling aspects of society. Guilds are both good and bad: they bring the benefit of quality control but come with substantial costs.
Somewhat repetitive, but well constructed. I think delving into the history of occupations is necessary to get the full picture of this. We can see the formation of guilds in the formation of two of the first nation-states: Egypt and India. The animal headed gods are likely symbols of the shamans who were the connection between the people of the separate tribes of the nation and the protector spirit of a species of animal. Thus there was, likely, in prehistoric Egypt, a tribe worshiping the jackal, another worshiping the lion, alligator, ibis, etc. When the Sahara began to dry out, these all migrated to the Nile Valley, where each tribe took on a hereditary occupation so as not to fall into warfare with one another. Thus, the tribe of the jackal became hereditary undertakers, the tribe of the falcon, aristocracy or royalty, etc. This was the process of tribes becoming castes.
After some time like this, people must have entered Egypt who did not come with a tribe. Others may not have been of the Egyptian nation at all. At the same time, new skills were needed in Egypt as certain resources were depleted and others were discovered, and technology changed. Also, as in common, some tribes probably dwindled. All these causes led to the beginning of apprenticeship. Originally, in Egypt, this was a form of adoption. We can see why when considering the above described origin of caste in tribes becoming civilized.
Guilds probably originated in castes which had become extinct while their services were still valuable.
While economic systems have changed markedly since ancient times, aste systems have grown into class systems, but retained most of their features. While the pecking order is different from the original, the business class, military class, and intelligentsia jockey for position while exploiting the working class and oppressing outcast minorities in all Indo-european societies. Your purely economic and social status arguments fail about guilds fail to include this sort of thing, so fail to show just how fundamental guilds are to modern Western societies and how unlikely they are to be reformed.