On Credentials, Trust, and Decentralization of Heuristics
The college degree is a signaling tool — but a blunt one. Like most blunt tools, it doesn’t do its job very well. Smaller, finer credentials are better at signaling skills and characteristics and are becoming easier than ever before.
The secret of higher education is that people don’t go to college for an education — they go for a credential.
Amid all of the hoopla over how disruptive new education delivery mechanisms like MOOCs are, people overlook the fact that they haven’t actually been that disruptive at all. So long as they don’t issue credentials that can actually be used to filter skilled candidates from unskilled candidates, they won’t disrupt much more than the YouTube video industry.
They needn’t be perfect signals or perfect credentials — they just need to be better than the next best thing. And the reigning champion of credentials is the college degree. With the influx of federally subsidized student loans into higher ed and for-profit colleges popping up to take advantage of these, the strength of this credential comes down. If everybody has a signaling tool, what can it really signal?
On one hand, it can signal that those who don’t have it are truly the bottom of the barrel. If all you need is a pulse to get into college and an ability to sign a student loan form, you really have to be a bad candidate to not have a credential. Or so the idea goes when you ignore edge cases.
On the other hand, what the credential signals is less significant. When anybody can go purchase a degree, even those who work hard over the course of four years to learn and grow are left hurt by those who just walk into a for-profit college.
You can test this secret by going to any group of college students and asking them whether or not they would still spend four years, a lot of money, and take time off from everything else to do what they are doing now — except that they wouldn’t get a degree when they’re done. Save the few who truly love academia, most would look at you like you are crazy.
The Dilemma of Heuristics
Any credential is just a heuristic — a way for somebody to make judgements quickly and shortcuts for decisions without having to spend a ton of time doing research about a candidate, their work, and their reliability.
Most heuristics come with an element of trust inherent within them. You outsource certain decisions to other groups, institutions, and individuals who have proven themselves capable.
Social trust works this way — if somebody has certain known standards for associating with people (i.e., as a friend, as an investor, as an advisor, etc.) — then others can choose to associate or unassociated quickly based on the trust they hold in other people.
College credentials work this way, too. If Stanford has certain standards for letting somebody in, people can trust Stanford’s decision-making much more quickly than they can trust their own. If a school has poor standards and those become known, then a college credential can work negatively, as well. (I can speak from experience in recruiting that I am less likely to spend time on certain campuses than others because of the caliber of student I meet.)
But how much trust can you wrap up in an institution? How blunt is this trust?
Trusting large credentialing mechanisms like colleges and universities for specific skill sets doesn’t make a lot of sense unless they’ve developed a reputation for a specific skill set. Trusting the category generally makes even less sense (i.e., hard and fast credential requirements for hiring decisions).
You may know that somebody has a certain level of competitiveness or ability to pay attention to details based on their admission to one college over another, but if you are trying to hire for a specific skill set, a general credential isn’t going to help as much as a finer, more focused credential.
Solution: Decentralizing Credentials — Less Blunt, More Fine
The solution is simple — further decentralizing skill-based credentials into the institutions that have the most trust associated with those fields.
This is obvious and already done for a lot of companies and the products they build. Salesforce or Oracle may certify people in their software, Amazon provides exams for IT professionals who want to work with Amazon Web Services, and sales consulting groups will certify people who meet their standards (i.e., Sandler sales system).
Rather than seeking a credential that includes a large bucket of general categories and soft skills, the finer credentialing service allows you to focus on seeking a credential that signals an understanding of a small, but focused bucket. Rather than spend four years, the candidate spends a few weeks. Rather than spend six figures, the candidate spends a few thousand (at most).
This isn’t anything new, but the accessibility to decentralized, skill-oriented credentials is higher now than ever. The delivery of knowledge is one side of the equation (i.e., being able to access classes in Salesforce, AWS, or Sandler without having to already have a job in the field and have your employer pay for it).
The other is the general credential marketplace.
These private, decentralized credentialing services have every incentive in the world to make sure that they don’t just give out certifications to anybody who comes asking for one. Their customers are paying for the trust others have in the organization and if people lose trust in the organization, customers will stop coming to them to earn that trust.
With blunt credentialing tools like colleges and universities, this is less-true. Admitting a few flukes into Penn State or UT Austin won’t sink the school with the long, established history it has. And even if it could, that really wouldn’t matter because the students are rarely the customers; rather, their parents or the government are the actual customers and they are mere inputs that allow the schools to get parental money or student loans.
If Amazon were to certify just anybody in their AWS Solutions Architect Associate exam, for example, then fewer people would be hired based on that certification because employers would figure out that it really didn’t mean much. If fewer people are hired based on that certification, fewer people will pursue the certification. If fewer people pursue the certification, employers will be more likely to move to competing services that have better training and certification.
We see this today as more and more employers are open to alternative solutions to the college degree. The blunt, general signal captured with the degree is less useful for them as they need specific skills and as more people have the diluted credential. The removal from marketplace incentives means that what people learn over those four years could easily be outdated — while the private credentialing services have every incentive to keep people in the loop as they update their technologies and strategies.
If one were to predict how the market will move in coming years, it would likely look like this:
— More companies offer their own certification and credentialing services as skills must be finely tuned with new stacks.
— Employers place a heavier emphasis on these than on generic credentials.
— Generic credentials do not go away, but the individual to whom they are attractive changes. Rather than be attractive to everybody on the job market, they only are for those who need them to get by.
Making predictions is rarely a winning business, but this is the direction of the market right now.
