NextFin News - The Harbor Human Capital Factor US Large Cap ETF (ticker: HAPI) reached $501.8 million in assets as of September 14, 2026, a milestone that has turned a behavioral-finance experiment into one of the most discussed factor products of the year. The fund does not screen for earnings, leverage, or revenue growth. It buys companies whose employees say they like working there. For a while, that worked: HAPI gained 30% in 2023 and rose 12.3% in the first quarter of 2024, ahead of the S&P 500's 11%. Then the trade stopped working. Over the year through September 11, the fund returned 17.31% at net asset value, versus 22.32% for the S&P 500.
The timing is pointed. David van Adelsberg, founding partner and CEO of Irrational Capital - the research firm that built the fund's scoring model - appeared on September 14 to explain why employee happiness should matter to shareholders. The premise is seductive in its simplicity: happy workers are more productive, more innovative, and less likely to quit, and those advantages should eventually show up in cash flows and stock prices. But the fund's own performance table now poses the harder question: is the Human Capital Factor a durable source of alpha, or was it a beta gift that arrived wearing a culture label?
What the Fund Actually Buys
HAPI tracks the Human Capital Factor Large Cap Index, which starts from the Solactive GBS United States 500 - effectively the 500 largest U.S. companies - and selects roughly 150 names using Irrational Capital's proprietary scoring methodology. The index is sponsored by CIBC World Markets and calculated by Solactive; Harbor Capital Advisors serves as the investment adviser. The fund launched in October 2022 and carries a net expense ratio of 0.35%.
The scoring itself uses no financial data at all. Irrational Capital evaluates companies across behavioral dimensions - trust and transparency, psychological safety, autonomy, motivation, purpose, compensation fairness, and leadership quality - and sorts them into buy, sell, and neutral buckets. Roughly the top 150 make the cut. The firm says it has collected more than two million survey responses and 13 million unique data points from over 4,000 companies, drawing on public sources such as employee-review sites and proprietary third-party surveys. The weighting of each dimension varies at every reconstitution, which means the model's definition of a good culture is itself a moving target.
"This is purely the perception of employees, and we've codified it into the human capital factor," van Adelsberg said in an interview.
After selection, the index applies modified market-cap weighting with sector neutrality. Individual positions are capped at 5% or five times their weight in the starting universe, whichever is lower. On each annual reconstitution, sector weights are reset to match the Solactive 500. If a sector does not have enough high-scoring companies to fill its target, the index inserts SPDR Select Sector ETFs as proxy holdings - an unusual design that acknowledges culture is not evenly distributed across industries.
The result looks familiar. Information technology accounted for 32.6% of the portfolio as of August 31, 2026, with communication services at 14.2%. The fund's largest positions - Amazon, Microsoft, Apple, NVIDIA, Alphabet, and Meta - are the same names that dominate the S&P 500, and the top ten holdings together account for 42.6% of assets as of the fund's most recent disclosure. In other words, HAPI's "culture" portfolio is concentrated in the companies that can afford generous pay, flexible schedules, and on-site amenities. The fund's own prospectus concedes the limitation in plain language: "There is no guarantee that the construction methodology will accurately assess a company to include or exclude it from the index."
The Performance Record, and the Overlap Problem
The early numbers were the fund's best marketing material. HAPI rose 30% in 2023, ahead of the S&P 500, and gained 12.3% in the first quarter of 2024 versus 11% for the benchmark. Since inception, the fund has returned 24.84% annualized at NAV, compared with 23.78% for the S&P 500 - a margin of about one percentage point a year.
But that margin has been shrinking, and recently it flipped. Over three years, HAPI has compounded at 20.39% annually versus 20.61% for the S&P 500 - effectively a tie. Over the most recent twelve months, the fund has lagged by five percentage points. The tracking difference versus the fund's own benchmark runs about 0.40%, consistent with the 0.35% fee plus trading costs, so fees alone do not explain the shortfall.
The overlap is the uncomfortable explanation. HAPI's holdings carry the fingerprints of every established factor that has worked over the past decade: large-cap growth, quality, momentum. Companies with high employee-satisfaction scores also tend to have strong balance sheets, stable margins, and pricing power - the classic Quality profile. When a "culture" portfolio looks this much like a Quality portfolio, the burden of proof shifts to the sponsor: is the Human Capital Factor a distinct source of alpha, or a new label on an old factor?
Independent analysis has flagged exactly this. HAPI shows higher sales and earnings growth than the S&P 500 and a nearly identical forward P/E - characteristics that overlap heavily with established factors. With fewer than four years of live data, attributing outperformance specifically to employee satisfaction rather than to Quality is a hard case to make. The honest verdict is that the record is not yet long enough to separate the signal from the style.
There is a second, more uncomfortable possibility: the factor may have worked precisely because it was small. When a strategy is niche, its signal is uncorrelated with the crowded trades. As assets grow - and HAPI has grown more than tenfold since launch - the rebalancing flows themselves begin to move prices, and the edge decays. A culture screen that selected 150 names when the fund held $50 million faces a different market when it holds $500 million. Size is the silent killer of factor returns, and HAPI is only just beginning to test it.
Why the Signal Might Still Be Real
There is a serious academic case on the other side. Organizational psychology has spent decades documenting the link between engagement and performance: employees who trust their managers and feel autonomous produce more, defect less, and generate more usable ideas. Turnover is expensive - recruiting, onboarding, and lost institutional knowledge can cost a substantial fraction of an employee's annual salary. If the market systematically underprices these effects because they sit outside the traditional financial statements, then a culture screen is not a gimmick; it is an arbitrage of an information gap.
Irrational Capital's model leans on that gap. By measuring the psychological relationship between employees and employers rather than the financial relationship between a company and its creditors, the firm argues it is capturing a variable that appears in no earnings release. Dan Ariely, the Duke University behavioral economist who co-founded the firm, put it directly in the fund's launch announcement: "Our work has clearly proven that corporate culture and employee motivation are deeply connected with financial performance."
The mechanism is plausible, and it is second-order in a way that matters. A culture score does not move a stock through sentiment alone. It works through retention, which lowers operating costs; through discretionary effort, which raises output per dollar of payroll; and through risk avoidance, because companies with high trust are less likely to produce the scandals, lawsuits, and regulatory blowups that destroy value. That chain - perception to behavior to cash flow to price - is the transmission mechanism that would make the factor durable rather than decorative.
There is also a valuation discipline built into the design that defenders understate. Because the model uses no financial data, it cannot chase expensive stocks on purpose; a company enters on culture alone and is then weighted by market cap with a 5% cap. That prevents the strategy from becoming a momentum fund in the purest sense - it can hold a high-flying name only if employees rate it highly, and it can hold a cheap, unloved name if the workforce believes in it. In theory, that constraint should produce a portfolio that looks less like the index over time, not more. So far, it has not.
The Concentration Nobody Talks About
There is a second risk that has little to do with culture. HAPI's growth has not been purely organic. The State of Wisconsin Investment Board was an early institutional investor in Harbor's human-capital ETF suite, committing roughly $362 million across the strategy family, according to reporting at the time - at one point a substantial majority of the suite's total assets. HAPI's own daily trading is thin: 257 shares changed hands on September 11, and the 30-day median bid/ask spread sits at 0.13%.
This matters because factor strategies are fragile when they are crowded by a single allocator. A large withdrawal would force redemptions at the fund level, which flow through to the underlying stocks regardless of their culture scores. In a strategy that markets itself on a proprietary, non-transparent signal, liquidity risk is not a footnote - it is a test of whether the factor can survive its own success.
The structural irony is sharp: a fund built to measure the stability that good culture provides is itself exposed to the instability of a single institutional flow. If the culture signal is real, it should be robust to the comings and goings of one pension allocator. If it is not, then the fund's $500 million in assets is less a verdict on human capital and more a bet on one investor's patience.
Cyclical or Structural: The Call
So is employee satisfaction a structural source of alpha, or a cyclical one that will mean-revert? The verdict has to be split, and the split is the story.
The structural leg is real: the link between engagement and productivity is grounded in decades of research, and it does not depend on the direction of the market. Companies that retain talent and avoid culture-driven scandals should compound value faster than those that do not, across cycles. That mechanism does not expire when the Fed moves or earnings slow. A firm that loses fewer engineers to a competitor, or avoids one reputational disaster, keeps value that would otherwise have leaked away.
But the investable expression of that insight - as currently constructed - is cyclical. HAPI's outperformance arrived almost perfectly alongside the mega-cap technology rally, and its recent underperformance has arrived alongside the rotation away from it. The fund neutralizes sector weights, but its stock selection within sectors tilts toward the winners of the last cycle. When the market's leadership rotates, the culture signal is tested against a backdrop where its largest holdings are no longer the market's favorites. That is when we learn whether "happy employees" is an independent factor or a proxy for momentum.
The cyclical call, then: a meaningful portion of the alpha generated since 2022 was a beta gift from the technology complex, and some of it is now reverting. The structural call: the underlying mechanism is real, and a purer, less concentrated implementation could capture it without the overlap. Investors should not confuse the two. A factor can be true in theory and disappointing in implementation, and both statements can be correct at the same time.
What Would Prove This Wrong
The strongest counter-thesis is straightforward: HAPI is a Quality factor in culture clothing, and its modest lifetime edge over the S&P 500 disappears once you control for the characteristics its holdings already share with established factor indexes. If, over the next two to three years, HAPI's risk-adjusted return falls in line with a plain Quality or low-volatility strategy after fees, the "human capital" label will have been branding rather than a distinct signal.
The falsifying signal is specific: if HAPI underperforms the S&P 500 by more than its 0.35% fee for two consecutive calendar years while its top-ten concentration remains above 40%, the culture thesis should be treated as unproven. Conversely, if the fund holds up or outperforms during a period when mega-cap technology lags the broader market, that would be the first real evidence that the factor stands on its own.
What to Watch
In the short term, watch flows and liquidity: a large institutional redemption would test the fund's trading mechanics more than its investment thesis. Over the medium term, watch the annual reconstitution - specifically whether the fund continues to hold the same technology names or actually rotates into cheaper, less obvious culture winners. Over the long term, watch the factor's performance through a full market cycle, because fewer than four years of data is a sample, not a verdict.
The beneficiaries, if the thesis holds, are patient investors willing to treat culture as a multi-year signal rather than a quarterly trade. The exposed are investors who buy HAPI thinking they are getting diversification, when what they are getting is a concentrated bet on the same companies they already own - with an extra 0.35% fee layered on top.
HAPI has done something genuinely useful: it has forced the market to price human capital as a factor instead of a footnote in the annual report. Whether that factor pays on its own, once the technology rally fades, is the question the next cycle will answer. The market is not paying for happiness; it is paying for the cash flows that happy employees may eventually produce - and those cash flows have not yet been tested in bad weather.
Explore more exclusive insights at nextfin.ai.
