Does It Matter What College You Go To? What an Education Researcher Says
Written By: Ryan Morrison.
Based on a Navigating Wealth conversation with Michael Horn.
2025 High-Net-Worth Professional Services Study
See what HNW households with an average net worth of $5M+ spend on advisors, estate planning, childcare, and education services — benchmarked across 200+ respondents.
The college decision has never cost more or felt less certain. Tuition at selective private schools now routinely exceeds $90,000 a year. Entry-level job postings have fallen sharply as AI takes over the work that used to go to new graduates. And the standard advice (aim for the most selective school your child can get into) has started to feel less like wisdom and more like an inherited reflex. Michael Horn has spent two decades researching how education systems work, and his answer to whether where you go to college matters is more specific, more useful, and more honest than most of what parents hear.
Whether where you go to college matters depends heavily on which tier of institution you are comparing. For highly selective schools, the network and credentialing advantages are real, durable, and not easily replicated elsewhere. Horn describes the advantage as getting two lottery tickets instead of one to be an outsized earner in the economy. For mid-tier schools, that case weakens considerably, and the research suggests what you do there tends to matter more than the name on the diploma. The AI era is accelerating both conclusions.
Key Takeaways
For highly selective schools, the network and credentialing advantages are real and durable — Horn frames it as two lottery tickets instead of one to be an outsized earner
The case for college prestige weakens significantly below the top tier; what you do at the school matters more than which school it is
74% of students who went to college without a deeper reason for being there dropped out or transferred; the why matters as much as the where
AI has inverted the expected payoff for some technical degrees; CS majors now carry roughly double the unemployment rate of art history majors
55-85% of jobs are filled through personal networks rather than applications, and that share is rising as AI screens out cold submissions
A structured gap year, with a real job, a service component, and a genuine question to answer, can provide the clarity that prevents the most expensive version of the college mistake
Table of Contents
Does College Prestige Matter and for Whom
For highly selective schools, prestige still delivers something real, but what it delivers is not the curriculum. Horn is direct on this point: the value of an elite college has always been the network and the credentialing function, not the content of the courses. Tad Fallows put it plainly in the conversation: before Khan Academy existed, a student could have gone to a public library and checked out a calculus textbook. The point of attending an elite college was never the subject matter.
What elite schools provide is a screening function and a tribe. When you gain admission to a highly selective institution, you join a pool of people who have passed the same filter. Half of Harvard graduates end up marrying someone who attended Harvard or another selective graduate program. Founders meet future co-founders. Analysts meet future managing directors. The network compounds over decades in ways that a strong academic program at a mid-tier school rarely replicates, not because the education is better, but because the people in the room with you are different.
Horn frames the investment case through the lens of Jeff Selingo, his co-host on the Future You podcast and author of the book Dream School. Selingo's argument is that attending a highly selective school gives a graduate roughly two lottery tickets instead of one to be an outsized earner in the economy. Horn's read: for a similarly priced alternative that is not as highly selective, two lottery tickets looks like pretty good odds. Research from the HEA Group found that obtaining a degree from a top-tier selective institution was associated with a 14 to 19% increase in median earnings compared to peers.
The network argument holds beyond earnings. The exclusive club function that elite colleges serve is not accidental and it is not going away. As Horn notes, humans have always sought ways to form tribes with recognized membership. Colleges have served that sorting role for centuries, shifting the criteria over time while maintaining the fundamental social architecture. That architecture is part of what parents are buying when they write the tuition check.
Watch the Full Conversation
This article draws on a Navigating Wealth conversation with Michael Horn, where we discuss whether elite college is still worth it, how AI is reshaping entry-level jobs, what mastery-based learning models are getting right, and when a gap year makes sense. Watch the full episode for the broader discussion.
Michael Horn is an education researcher, author, and co-host of the Future You podcast with Jeff Selingo. He co-authored Job Moves, a guide to navigating career transitions with purpose, and writes the Future of Education Substack at michaelbhorn.com. His work focuses on how education systems can be redesigned to better serve students and the workforce, including the role of technology, mastery-based learning, and school-workforce integration. In this conversation, he explains why the college decision is more nuanced than most parents are told, and why the AI era is changing the calculus in ways that make preparation more important than selection.
Why the Case Weakens as You Move Down the Selectivity List
Below the top tier of selective schools, whether college is a good investment becomes more complicated, and Horn's position shifts from "yes, clearly" to "it depends on what the student does there." Selingo's argument, which Horn endorses, is that many colleges produce strong positive ROI, and that an institution's selectivity is far less predictive of outcomes than most families assume once you move below the top tier.
The most important variable Horn names is the student's reason for being there. His research found that 74% of students who went to college primarily because it was expected of them, without a genuine reason for being there, dropped out or transferred. The experience of every college student includes at least one hard stretch. Students who had a deeper reason for attending had something to hold onto during that stretch. Students who were there because it was the next thing you do after high school frequently did not. That is not primarily a story about college quality. It is a story about preparation and intention.
The data on the return on a college degree remains broadly positive. The Federal Reserve Bank of New York estimates the return at roughly 12.5% annually, a rate that outperforms most long-term investments. But that average conceals significant variation by school, major, and completion. A student who attends a mid-tier school, finishes with a purposeful major, and graduates with manageable debt is likely to see strong returns. A student who enrolls without a clear reason, changes majors twice, and takes six years to finish is not operating in the same investment equation.
Horn adds a counterintuitive finding worth attention: being a big fish in a smaller pond often produces better outcomes than being a small fish in a more selective one. STEM majors who attend top schools like Stanford sometimes abandon their major entirely because the competition is too intense relative to their preparation. The same student who would have thrived as a standout at a less selective school can find that the prestige environment works against them. For parents weighing a full-cost offer from a top-20 school against a merit scholarship at a strong regional school, that dynamic is worth factoring in. Understanding how HNW families think about preparing the next generation involves exactly these kinds of non-obvious tradeoffs.
How the AI Era Is Changing What a College Degree Is For
The AI era has not made college less valuable on average, but it has changed which preparation matters inside the college years. The labor market signal is striking. Entry-level and internship job postings fell 15% in 2025, while applications per posting rose 26%. The unemployment rate for recent college graduates reached 5.8% in early 2025, higher than the national average and higher than what high school graduates faced. These are not the numbers of a system working as intended for new entrants.
The specific inversion Horn names is worth underscoring. CS majors coming out of college now carry roughly double the unemployment rate of art history majors. For a generation told relentlessly that technical skills were the path to job security, that reversal reflects something real: AI has absorbed exactly the routine technical work that used to be the entry point for CS graduates. The skills that remain genuinely hard to automate are the ones that liberal arts degrees tend to develop despite their perceived impracticality: writing clearly, reasoning across domains, navigating ambiguity, managing relationships.
Horn's recommendation for students is to pair technical literacy with the humanities rather than choose between them. The market is pointing toward people who can use AI as a tool while applying judgment, communication, and domain knowledge that AI cannot easily replicate. That combination is more valuable than narrow technical depth alone.
What has not changed, and what Horn argues will become more valuable rather than less, is the network. Between 55 and 85% of jobs are filled through personal connections rather than open applications. As AI processes cold submissions more efficiently, the personal introduction becomes the reliable path into opportunities. Sending a thousand applications is no longer a volume strategy. Knowing ten people who know the right other people is. The college environment that builds those relationships early tends to pay back over a career in ways that never appear on a job board. This is part of what founders who have navigated multiple career transitions understand from experience.
Beyond Wealth Newsletter
Weekly perspectives on money, meaning, and the decisions that come after the financial ones get easier. Read by founders, executives, and investors navigating the same questions covered in this post.
What Mastery-Based Learning and New School Models Are Getting Right
Real assets (real estate, commodities, infrastructure, timberland) hold value through currency depreciation because their worth is grounded in something physical rather than a promise denominated in paper. When a government inflates its way out of a debt problem, the nominal price of real assets typically rises to preserve their purchasing power. That is the structural logic behind holding them, and it is distinct from the momentum or yield arguments that often dominate the discussion.
The historical record supports this. Precious metals and international equities are among the only asset classes to consistently deliver positive real returns for dollar-denominated investors during sustained eras of dollar depreciation, according to Barclays Private Bank research. Cash and nominal bonds, the default defensive positions in many portfolios, have historically provided limited protection in these environments because their real value erodes alongside the currency.
Real estate deserves specific attention within this category. Property generates income through rents that can be revised upward as replacement costs and general price levels rise. The asset itself is a store of value that does not depend on any single government's willingness to maintain purchasing power. For investors thinking about how HNW investors allocate to real estate as part of a broader portfolio, the inflation hedge argument is one of several compounding reasons the asset class tends to hold a meaningful share of HNW portfolios across market cycles.
Commodities operate differently. Most are priced globally in dollars, which means a weaker dollar tends to push their dollar-denominated prices higher even when underlying demand has not changed. That dynamic benefits commodity-exposed positions but also illustrates how dollar depreciation operates as a tax on domestic purchasing power rather than a generalized wealth event. Private credit as a complement to real asset exposure is worth examining in this context: floating-rate private credit denominated in foreign currencies adds a layer of non-dollar income that is distinct from equity exposure while still keeping the portfolio productive.
The broader principle Altman articulates is that a portfolio heavily concentrated in paper instruments (stocks, bonds, cash, and funds denominated in dollars) is implicitly long on the US government's willingness and ability to maintain purchasing power. Real assets represent a partial hedge against the failure of that implicit assumption.
The Gap Year Question: When It Works and When It Does Not
A structured gap year is one of the more underused tools for avoiding the most expensive version of the college mistake. The research is supportive: students who take gap years earn higher GPAs in college than their pre-gap academic profiles would predict. They arrive more motivated, with a clearer sense of why they are there, and they are less likely to be among the 74% who drop out or transfer because they lacked a genuine reason to be enrolled.
Horn's framing of what makes a gap year valuable is specific enough to be actionable. The gap year should not be aimless travel. It should include a real job (something with actual responsibilities and real consequences), a service or volunteer component, and some structure for figuring out what genuinely drives the student's interest. The combination of earning, contributing, and exploring tends to produce the kind of clarity that prevents a student from arriving at college and treating the first two years as a continuation of adolescence at high cost.
Tad pushed back in the conversation, noting that Horn sounded like no fun as a parent. Horn's response was precise: he would provide a budget for the backpacking component, but there are caps on everything. A gap year as a sabbatical from thinking about what you want to do is a different thing from a gap year as a structured exploration of exactly that question.
For Long Angle members whose children are college-age or approaching it, the practical implication is that the gap year conversation is worth having before the college application process becomes the organizing frame. A student who takes a meaningful gap year and arrives at college with a genuine reason for being there is likely to get more out of the experience, regardless of which school they attend.
Frequently Asked Questions
Does it matter what college you go to for your career?
For highly selective schools, yes. The network and credentialing advantages are real and persist well beyond the first job. For mid-tier schools, the research suggests the institution's name matters considerably less than what the student does there, how they build relationships, and whether they finish with a clear sense of direction. The gap between tiers is large. The gap within the mid-tier is smaller than most families assume.
Is an Ivy League degree worth it?
If you get in and the cost is manageable, Horn's view is that it is worth attending. The social network, the screening function, and the access to outsized earners in the same cohort are advantages that compound over decades. Selingo frames it as two lottery tickets instead of one to land in the top tier of earners. For most families, that is a meaningful advantage over a similarly priced alternative.
Is college a good investment in the age of AI?
On average, yes. The Federal Reserve Bank of New York estimates the return at approximately 12.5% annually, which outperforms most long-term investments. But the average conceals large variation. The students most at risk are those who enter without a clear reason for being there, who pursue narrow technical skills without broader analytical development, and who do not build the personal networks that fill the majority of good jobs.
What are the benefits of a gap year before college?
Students who take structured gap years earn higher GPAs in college than their pre-gap academic records would predict. They arrive with more motivation, greater clarity about their goals, and a lower probability of dropping out or transferring. The key distinction Horn draws is between a gap year with structure (real work, service, and genuine exploration) and one that functions as extended avoidance of the question of what the student wants to do.
What is mastery-based learning and why does it matter?
Mastery-based learning holds that students should not advance to new material until they have demonstrated genuine understanding of the current content, typically at a 90%+ threshold rather than the 70-80% common in commercial educational software. The Alpha School model, Acton Academy, and Summit Public Schools all operate on some version of this principle. The result is that students move through academic content faster and with better retention, freeing time for the experiential and project-based learning that develops the interpersonal skills AI cannot easily replace.
Does college prestige matter for jobs?
For the first job, selective school prestige opens doors that are harder to open otherwise. For subsequent roles, the personal network matters more than the credential. Half of jobs are filled through personal connections, not applications. The college that builds a denser, more useful network tends to keep paying dividends. For elite schools, that is part of the value proposition. For schools further down the selectivity list, it is one more reason to invest in relationships rather than relying on the diploma alone.
How should parents prepare kids for an AI job market?
Horn's framework has three components. First, prioritize real work experience over additional academic credentials — the market now expects more demonstrated capability, not more resume-padding. Second, resist the narrative that technical skills alone guarantee security; the skills hardest to automate are communication, judgment, and relationship-building. Third, help young people identify what genuinely drives them, because students who know why they are doing something tend to outperform those who are chasing a credential for its own sake.
Final Thoughts
The college question has always been more about network and credentialing than curriculum. Horn's research makes that explicit rather than leaving it as a polite fiction. The Ivy League's academic content was never the product. What you were buying was the screening process and the social architecture it created. AI accelerates that reality rather than changing it. The students who will get the most out of college, at any selectivity level, are the ones who understand that and use the years accordingly: building relationships, seeking mentors, finding real work, and arriving with a genuine reason for being there.
For parents making these decisions alongside their children, the useful reframe is from "which school" to "does my child have a real reason to go, and are they prepared to use the time well." The school matters at the top of the selectivity distribution. Below that threshold, the preparation matters more than the selection.
The network Horn describes as the real product of elite college — vetted peers, candid conversations, people who vouch for you by name — does not stop being valuable after graduation.
Long Angle is a vetted community of 8,000+ high-net-worth founders, executives, and investors where members compare notes on the decisions that matter most: portfolio construction, business exits, estate planning, and how to prepare kids for a world that looks nothing like the one we graduated into.
Resources Mentioned
Dream School by Jeff Selingo — the case for rethinking the college selection process
Job Moves by Michael Horn — navigating career transitions with purpose in a changing economy
Future You podcast — Michael Horn and Jeff Selingo on education, careers, and what comes next
Michael Horn's Substack: Future of Education at michaelbhorn.com
Acton Academy — mastery-based, project-driven school model, founded in Austin
Alpha School — two-hour academic model with AI-adaptive tools and experiential learning blocks
Summit Public Schools — personalized learning time plus project-based curriculum, California
Watch or Listen
Watch on YouTube · Listen on Spotify · Apple Podcasts · Amazon Music
