Playing cat-and-mouse with higher education

By Jam Magdaleno and Cesar Ilao III

(Part 1)

IN RECENT WEEKS, the upper chamber’s political ruckus has once again demonstrated its unrivaled ability to divert public attention from the more mundane afflictions of the republic. What’s lost in the commotion was the news of the proposed reframing of the General Education (GE) curriculum.

The proposal, first floated by the Commission on Higher Education (CHED) in April, sought to trim the existing 36-unit GE curriculum down to 18 beginning School Year 2026-2027. Currently, CHED’s GE policy mandates a uniform 36-unit curriculum — anchored on eight core subjects — that all Philippine universities must include in every bachelor’s degree, with only limited flexibility in electives and no ability to drop core requirements.

Reframing the GE curriculum, CHED said, will better align university outcomes with labor market realities increasingly shaped by AI. This effort, of course, comes at a time when college graduates now confront diminishing employability at home and abroad.

Predictably, however, the proposal drew immediate fire from academics and the public, forcing the CHED to delay the implementation of the proposed reframing to 2028. CHED Chairperson Shirley Agrupis said that the postponement will “give us time to analyze and study the different manifestations well.” Setting aside, for the moment, the substantive merits of the position papers raised by critics, CHED’s response raises a more fundamental question about the peculiar logic of education policymaking under conditions of rapid technological change.

If, as the CHED itself argues, rapidly evolving technological change now demands urgent curricular realignment to meet shifting labor market needs, then a two-year postponement appears to undercut the very rationale of the reform. Urgency, after all, is a perennial enemy of Philippine bureaucracy. But if one counterargues that postponement allows regulators to gather more evidence and refine the proposal further, then the question naturally arises: Why stop at two years? Why not three, five, or 10? After all, more waiting invariably promises more data, more consultations, and more refinement.

To state the obvious elephant in the room, CHED’s presumed ability to predict, future-proof, and centrally plan around the labor market is, at best, deeply unreliable and, at worst, an impossibility. The very nature of labor markets is shaped by creative destruction. New technologies continually displace old industries, jobs, and skills. What we deem valuable today may become obsolete tomorrow, and new forms of work emerge that no all-intelligent regulator could have anticipated beforehand. Most of our readers would probably be unfamiliar with the occupation of the “knocker-ups” once common in industrial Britain. Before alarm clocks became widespread, workers were hired to tap on windows with sticks to wake laborers for work. Few observers at the time could have predicted that a simple technological innovation would render an entire occupation obsolete.

The unpredictable nature of creative destruction is precisely why the notion of government-directed “labor-market forecasting” deserves far more skepticism than it typically receives.

Many economists have long observed that labor-market forecasting is notoriously prone to large errors because it rests on rigid assumptions about an ever-changing world filled with technological disruption and occupational substitution. Among those who have emphasized this problem is Nobel laureate Philippe Aghion. In his book The Power of Creative Destruction, Aghion explains how transformative technologies such as the steam engine, electricity, the internet, and now artificial intelligence (AI) belong to what economists call “General Purpose Technologies” (GPTs). These technologies are stubbornly resistant to centralized prediction because they evolve rapidly, are applicable across a wide range of industries, generate complementary innovations, and have the ability to reorganize the structure of economic activity itself. To illustrate the idea further, consider two uncomfortable realities: First, many of the industries in which the Philippines presently enjoys a comparative advantage (such as BPO/IT-BPM, customer service, administrative work/virtual assistance, seafaring labor, and segments of electronics manufacturing) are precisely those most exposed to technological disruption from AI.

Second, according to the recent EDCOM II report, curriculum revisions within CHED take, on average, 11 years to materialize. By the time its meticulous reform package reaches schools, the labor landscape it was designed to address no longer exists.

The fatal conceit in CHED’s approach to the whole matter that nobody seems to have pointed out lies in treating the labor market as an engineering puzzle solvable by a committee, rather than an unpredictable landscape that can only be navigated through adaptability. The allure of central planning, despite our bureaucracies’ solid record of moving at a glacial pace, blinds us from its inability to predict and plan education around future labor market conditions and rapid technological disruption.

We have discussed elsewhere the so-called “knowledge problem” in the context of primary and secondary school vouchers, but the logic applies equally here. No central bureaucracy, least of all one operating at the tempo of the Philippine state, can adequately aggregate the dispersed and rapidly changing information embedded in millions of uncoordinated decisions made by parents, students, teachers, administrators, employers, technologists, and others. Even international evidence points to the frailties of long-range labor-market forecasting. The IMF, examining decades of multi-country forecasting data across advanced economies, finds that labor-market projections are repeatedly revised in response to shocks and structural changes. The countries persisted in committing forecast errors over time, particularly during periods of technological transformation. In practice, policymakers often adjust their expectations ex post rather than accurately anticipating them beforehand. The future is often explained retrospectively.

But even if one grants, for argument’s sake, that the government can overcome the herculean task of forecasting future labor demand, this still does not account for the political economy of such reform. The current GE framework has been in place for nearly a decade. Unsurprisingly, entrenched interests have inevitably formed around it, which considerably raises the transaction costs of any attempt at reform. Educators who depend on the framework for their livelihood will fight tooth and nail for its retention.

Why does this matter? CHED’s arbitrary power over the GE framework — and the interests revolving around its maintenance — creates a “cumulative effect” on potential reforms. Regardless of present labor-market realities, implementing a new framework would require reversing course on the institutional arrangements that the previous GE system has already entrenched over the years. As a result, reframing the current GE curriculum becomes less likely to be debated on purely intellectual or scientific grounds, and more likely to be shaped by conflicting interests, all because the authority to impose a specific number of GEs rests, arbitrarily, with CHED. Such are the inherent dangers of interventionism. Now that we’ve stated these critiques, how should we reconcile the supposed mismatch between a GE curriculum that is increasingly misaligned with contemporary labor-market realities, and the inherent limitations of CHED in accurately forecasting those same labor markets?

(To be continued.)

Jam Magdaleno is the head of Information and Communications at the Foundation for Economic Freedom (FEF) and is an Asia Freedom fellow at the London School of Economics and Political Science (LSE) and King’s College London. Cesar Ilao III is a researcher and communications specialist for the FEF. He is a lecturer at the University of the Philippines and was formerly a researcher at Monash University, Australia.