Baxnet Ideas · Founder note
The Safest Person to Back May Be the One Who Needs Backing Least
TL;DR: The person with the clearest route to success may be the easiest investment to defend. They may also be the person whose future changes least because the money arrives.
Investing in People — Part Six. This article follows The Fund Has a Portfolio. You Have One Life., which looks at the individual’s concentrated risk even when the finance behaves fairly. If you are new to the series, begin with Everyone Is Investable. Nobody Should Be Ownable.. Part two asks how unconventional ability becomes credible, part three asks what the money should buy, and part four asks who absorbs financial loss.
Two applications reach the final meeting.
The first comes from somebody with a stable professional job, strong qualifications and an offer for a recognised course. Her employer has already agreed to reduce her hours. The likely next role has a published salary range, and several people with a similar background have made the same move. She needs help with fees and living costs. The case is tidy.
The second comes from a warehouse team leader who has spent the last year improving the way shifts and stock are organised. He has built a few small tools, become the person colleagues ask when a process fails and started learning more technical work at home. He wants one protected day each week to test whether he could move into logistics systems. There is evidence in the application, but the route is irregular. Nobody can point to a cohort of people who have already taken it.
Both could make good use of the money, and both may genuinely need it. The first applicant is approved.
A year later, she finishes the course and moves into the expected role. The fund records a successful outcome. Its selection process looks sensible because it was sensible. It found a capable person, backed a credible plan and produced a return that can be explained to an investment committee in one slide.
There is just one awkward question left: what would she have done if the fund had said no?
The outcome can be good and still tell us very little
Perhaps she would have delayed the course for six months. Perhaps her employer would have paid more, or she would have borrowed from family. Perhaps she would have chosen a cheaper route and reached the same job a year later. The funding may have made the transition faster, less stressful and more humane. That matters.
It is still different from making an otherwise unavailable transition possible.
Public policy has a dry word for this distinction: additionality. The UK Treasury’s Green Book separates outcomes caused by an intervention from “deadweight”, meaning outcomes that would have happened without it. The language is meant for government appraisal, not for deciding who should receive career finance, but the question travels rather well: what changed because the intervention existed?
Development investors use a related test. The International Finance Corporation describes additionality as making a contribution beyond what the market already provides, rather than displacing finance that was already available. The point is to direct scarce support where it adds something.
Applied to people, this does not mean the safer applicant is undeserving. It means a fund should know whether it is predicting a good outcome or causing one, and be clear about which achievement it is claiming.
Selection can quietly remove the reason for the fund
Imagine the investment committee refining its process over several years. It learns which qualifications correlate with completion, which employers provide dependable support and which career moves produce the most predictable earnings gains. Every unsuccessful case makes the filters a little tighter.
This is normal risk management. The problem appears when a market created to find overlooked potential becomes exceptionally good at identifying people who were visible already.
The warehouse team leader brings more uncertainty. His ability is distributed across small pieces of evidence: the scheduling problem he fixed, the colleague he taught, the evening course he kept returning to, the simple tool that his shift still uses. His next role has several possible names and no obvious salary ladder. Each detail may be positive, yet the whole case remains harder to defend.
That difficulty is partly the point. If ordinary lenders, employers and education systems already knew how to finance his transition on reasonable terms, he would not need a new market for human potential.
There is a broader pattern here. OECD work on adult learning finds that participation remains highly unequal: highly qualified workers are more likely to receive training, while older workers, people with lower skills and those in precarious jobs remain under-represented. Cost and time are among the barriers. The people who could benefit most are often the least likely to participate.
A selection system can reproduce that pattern without ever using education or job status as an explicit gate. Ask only for the evidence that successful conventional applicants tend to possess, and the gate rebuilds itself.
Better evidence does not decide what the fund is for
This is where I think personal intelligence can help, but only up to a point.
A person-controlled record could make unconventional capability easier to see. Instead of reducing the warehouse team leader to a job title and a self-written claim about being “innovative”, it could help him assemble the history behind the claim: problems solved over time, work other people began to depend on, skills practised consistently and the conditions in which he did his best work.
That is useful evidence. It could make a strange-looking application more legible to an employer, lender or investor.
It cannot decide the institution’s objective.
If the fund is rewarded only for predictable financial returns, richer evidence may simply make exclusion more precise. It will become better at ranking people by confidence, then congratulate itself for using more personal data to reach the same safe part of the market.
The boundary matters here. A personal intelligence engine should help the individual present relevant, inspectable evidence. It should not become a permanent feed of private messages, location history and relationships into an investor’s scoring system. Nor should it produce one durable potential score that follows somebody after a rejection.
Evidence can widen what a committee is able to recognise. The mandate determines what it chooses to value.
A fund has to choose some additionality on purpose
A serious fund still needs returns. If it backs only the hardest-to-predict cases, ignores evidence and treats uncertainty as a virtue in itself, it will not widen access for long. It will run out of money.
Preferring riskier people would be as crude as rejecting them. A workable fund would distinguish uncertainty worth investigating from a weak proposal, then build a portfolio that can afford to investigate some of it.
That might mean reserving part of the fund for applicants whose opportunity is credible but not already financeable elsewhere. It could mean starting with smaller amounts that buy a work sample, a placement or one day a week, rather than demanding enough confidence for a complete career change on day one. Public, employer or philanthropic capital could absorb some of the extra uncertainty where the wider benefit is real but the private return is not yet obvious.
The reporting would need to change as well. Completion rates and earnings gains tell the fund what happened to the people it selected. They do not show whether the fund altered those futures. A useful review would also ask what alternative routes were realistically available, whether the support accelerated an outcome, whether it improved its quality and whether it reached people other providers would have declined.
None of this produces a perfect counterfactual. We cannot observe the same person living an unfunded version of the year beside the funded one. The judgement will always contain uncertainty.
That is not an excuse to avoid the question. It is a reason to be honest about it.
The rejected application never appears in the success rate
At the next annual meeting, the first applicant’s result is on the page. She passed the course, moved roles and earned more. It is a genuinely good outcome, and she may feel that the funding changed the texture of a difficult year even if she would eventually have found another route.
The warehouse team leader is absent from the report. He was never part of the portfolio, so his result cannot damage its performance. Perhaps he kept learning. Perhaps the day job expanded into the time he had been using. Perhaps the small tool remained a useful curiosity and nothing more.
The fund can prove that it chose people who succeeded, but it cannot point to the opportunity it never allowed to become an outcome.
A market for investing in people should be judged by more than how accurately it spots the futures already coming into view. Its harder purpose is to help a credible future arrive because somebody chose to back it.