There is a belief about family businesses that is so widespread it rarely needs saying out loud: the business exists, at least in part, to give every relative a job. A nephew finishes school with no plan, so he joins...
There is a belief about family businesses that is so widespread it rarely needs saying out loud: the business exists, at least in part, to give every relative a job. A nephew finishes school with no plan, so he joins the shop. A cousin loses her position in town, so a role is found. An uncle needs something to do, so he becomes a manager of something. Each hire is an act of love, and refusing any one of them feels like a betrayal of what the family business is for.
The belief deserves a fair hearing, because it carries real values: loyalty, obligation, the refusal to let kin fall. But it also carries a testable claim, that a business can absorb relatives without limit and remain a business. That claim has been tested, at scale, by some of the most careful research in empirical economics. The results are not close.
Start with the largest measurement exercise ever conducted on how firms are actually run. Nick Bloom and John Van Reenen built a survey instrument that scores companies on concrete management practices, monitoring, targets, and people management, and deployed it on 732 medium-sized manufacturers across the United States, France, Germany, and the United Kingdom, later expanding into the World Management Survey covering thousands of firms worldwide. Their central finding was that these scores are strongly associated with productivity, profitability, and survival. Their second finding is the one that matters here. In their own words, poor management practices are more prevalent when product market competition is weak and when "family-owned firms pass management control down to the eldest sons (primo geniture)" (Bloom and Van Reenen, CEP Discussion Paper 716, 2006, published in the Quarterly Journal of Economics, 2007).
Read that carefully, because it is precise. Family ownership by itself was not the disaster. The disaster was allocating the top job by birth order rather than by ability. When the firm is inherited like a bed rather than contested like a title, management quality sinks toward the bottom of the distribution. Their later review in the Journal of Economic Perspectives lists family ownership among the main reasons badly managed firms survive at all (Bloom and Van Reenen, 2010).
Management scores are one thing. Money is another, and the money evidence is stronger still.
Morten Bennedsen, Kasper Meisner Nielsen, Francisco Perez-Gonzalez, and Daniel Wolfenzon studied CEO successions across Danish firms, using a clever natural experiment: the gender of the departing CEO's first-born child, which shifts the odds of a family succession but has nothing to do with the firm's quality. Their finding: family successions cause operating profitability on assets to fall by at least four percentage points around the transition, with the damage concentrated in fast-growing industries, industries that need highly skilled labour, and larger firms (Bennedsen et al., Quarterly Journal of Economics, 2007). Professional, outside CEOs, they conclude, provide extremely valuable services to the organisations they head.
Perez-Gonzalez found the same pattern in the United States: firms that promoted family-related CEOs underperformed on operating profitability and market-to-book ratios compared with firms that hired outside, and the underperformance was concentrated where the family appointee lacked strong educational credentials, which points at the mechanism. The problem is not the surname. The problem is drawing leadership from a talent pool of one family instead of a talent pool of everyone (Perez-Gonzalez, American Economic Review, 2006).
These studies target the CEO seat, but the arithmetic they expose runs straight down the payroll. If choosing one leader from a pool of relatives instead of a pool of candidates costs four points of return on assets, consider what it costs to fill ten positions that way.
You do not need econometrics to observe the machinery. A payroll full of untouchable relatives is a cost structure your competitor does not carry.
The competitor across the road hires the best cashier available and replaces her if she underperforms. Your shop carries the cousin who cannot be fired, at a salary that was never benchmarked, in a role that was invented for him. The competitor's manager can discipline anyone on the floor. Your manager cannot touch the founder's nephew, and every non-family employee watches that fact and updates their effort accordingly. The competitor promotes on results, so ambitious outsiders join and stay. Your best non-family staff can see the ceiling above their heads, wearing the family's face, and they leave. Each individual kindness is small. Compounded across a payroll and a decade, it is the difference between the firm that buries its rivals and the firm that becomes the family's most expensive charity, right up until it can no longer afford to be.
Fairness requires the counter-evidence, because family employment is not purely a tax.
Craig Aronoff and John Ward, who spent careers advising business families, document genuine advantages that family employees can bring: deeper trust, lower agency costs, longer horizons. A daughter who grew up on the business does not steal from the till, does not need her incentives engineered, and thinks in decades rather than quarters. There is also a fair critique of the research itself: the Danish evidence comes from a rich economy with a welfare state, while in economies without safety nets a family job can be rational insurance, the only social protection on offer. And management scores measure firm performance, not family welfare. A family may knowingly accept a less profitable firm that holds the family together, and that is a value judgment economics cannot overrule.
But notice what the counter-evidence actually supports. It supports employing capable relatives who are held to standards, where the trust advantage is real precisely because performance is real. Aronoff and Ward themselves warn that treating every family member equally regardless of contribution is what destroys cousin-generation firms. None of the evidence, on either side, supports the actual myth: unconditional employment for every relative who needs a job. The trust dividend is paid by competent family members. The incompetent ones spend it.
The escape from the myth is not coldness. It is separating two things the myth fuses together: the duty to support family, which is real, and the method of a payroll job, which is only one method and often the worst one.
First, adopt a family employment policy before anyone needs a job: family members qualify and work outside first, they are hired only into real vacancies at market pay, they receive genuine performance reviews from someone who is not their parent, and underperformance carries the same consequences as for anyone else, with the explicit written promise that leaving the payroll never means leaving the family. Rules written while every candidate is hypothetical feel like rules. Rules written after a niece applies feel like verdicts on the niece.
Second, route support through channels that do not sit on the payroll. A relative can be supported with dividends if they are an owner, and ownership without employment is a perfectly honourable position. They can be supported through a family bank that lends on terms for education, tools, or a business of their own, building their capacity instead of renting their presence. They can be supported with school fees and training, the one transfer that compounds. Every one of these channels delivers love in full. None of them installs an unmanageable person between your managers and your customers.
That is the real meaning of the evidence. Bloom and Van Reenen did not find that families ruin businesses. They found that businesses run as employment schemes get run like employment schemes. The family business can be the engine that funds the family's rise for three generations, or it can be the vehicle that carries everyone downhill together in comfortable seats. It cannot be both.
Put one item on the agenda of your next family council: how do we help relatives, and is the payroll the only tool we own? Draft the employment policy while no one is applying. List the support channels you will offer instead of invented jobs: dividends, the family bank, education funding. Then hold the line at the next funeral, the next graduation, the next quiet request, because the myth does not return with arguments. It returns with a nephew, and a chair that would be so easy to fill.