Credit transfer between institutions

How can a student carry credits between universities that never agreed on a curriculum?

A student finishes two years at one university and moves to another in a different country. The receiving registrar has never seen the syllabus of a single course she took, has no contact at her old department, and could not say what was on her examinations. Yet a fortnight later she is enrolled in third year, with two years of credit accepted.

Nobody agreed a curriculum. The two institutions may have entirely different traditions, term lengths and assessment styles. So what exactly was transferred, and what did the receiving institution rely on when it decided to trust it?

Begin by asking what the alternative would cost.

Imagine transfer required agreement on content. Then each pair of universities would have to negotiate a mapping between their courses, and the number of such negotiations grows with the square of the number of institutions. A few thousand universities give millions of pairs, each needing periodic revision as curricula change. This does not merely take effort; it is a problem of a size nobody can pay for. And it would fail anyway, because the whole point of separate institutions is that they teach different things.

So the working arrangement cannot be agreement about content. What else could it be?

Notice what a receiving registrar actually looks at. Not the syllabus. A transcript listing course titles, grades, and a number of credits — and the crucial move is in that number. A credit does not measure what was learned. It measures roughly how much work it took: nominal hours of study. The European system is explicit, defining a full academic year as 60 ECTS credits and one credit as some 25 to 30 hours of student workload. The American credit hour descends from the Carnegie Unit of 1906, which standardised secondary schooling by time in class, and it keeps that lineage — an accounting of hours, not of understanding.

That looks at first like a defect, and it is worth sitting with the discomfort. A unit that cannot tell you whether the student learned anything seems to be measuring the wrong thing entirely. But ask why that particular unit is the one everybody adopted, and the answer is that its thinness is exactly what made it adoptable. Two universities that agree on nothing about pedagogy can still agree that a year is a year. A standard everyone can accept without changing anything they do gets adopted; a standard that demands curricular concessions does not. So the system optimised for the property that lets it spread.

But a unit alone is not enough. Anyone can print a transcript. What makes the number believable?

This is the second half, and it is a different move. The receiving institution does not verify the courses; it verifies the institution, by checking that it is recognised by an accrediting body the receiver already trusts. Trust is established once, at the level of the organisation, and then extends to everything the organisation certifies. That is what collapses the impossible problem down to a manageable one: instead of every pair of universities negotiating, each university establishes a relationship with a small number of recognised accreditors, and any two that share a reference can transact without ever meeting.

Both halves are the same kind of thing — a shared reference nobody is compelled to use but everybody benefits from using once others do. That is what makes this a coordination problem rather than a bargaining one. There is no conflict of interest: every institution would rather transfer worked than not. What they lacked was not willingness but a common reference, whose value rises with how many others adopt it. Hence the Bologna Process, launched in 1999, spent its energy on comparable degree cycles, a common credit unit and a standard diploma supplement rather than on any shared syllabus. It built the reference, not the curriculum.

And it explains where the arrangement frays. Being deliberately thin, the standard certifies effort and provenance but not fit — and fit is what a degree programme needs. So the receiving institution keeps discretion over whether a credit counts toward its own requirements, and that discretion is where transferred credit quietly dies. It is not marginal: a United States federal review of national transfer data found students losing on the order of two-fifths of their credits. Coordination makes credits comparable and still fails to make them count, which is why bilateral articulation agreements persist alongside the standard — pairwise negotiation reintroduced exactly where the cheap universal reference runs out.

Three refinements.

First, the credit is a unit of account, not of achievement. It says how much study the course represented; the grade beside it carries the quality signal, and even that is only comparable through a shared understanding of what grades mean. Treating credits as interchangeable tokens of learning is the misreading that makes their frequent rejection feel arbitrary rather than predictable.

Second, no one designed this. The Carnegie Unit was created for another purpose, accreditation grew out of institutional self-regulation, and the transcript is an administrative record that acquired a second life as a portable document. The system is an accumulation of conventions that turned out to interlock — typical of coordination standards, unusual in engineered ones.

Third, thin standards have a cost as well as a virtue. Measuring hours rather than outcomes means a course can satisfy the unit without teaching much, and the reform movements that try to replace credit hours with demonstrated competencies are attacking exactly that. The difficulty they run into is the one this whole answer turns on: a richer standard is harder to agree, and a standard nobody adopts coordinates nothing.

• Competency-based credentials, and why a richer standard is harder to coordinate on.
• How micro-credentials and digital badges try to solve portability without an accreditor.
• Articulation agreements: where bilateral bargaining survives inside a universal standard.

Every term the collection defines is gathered in the glossary.