From Phlebas to IKEA

The millennial survival and dominance of patrimonial firms.

amphora label vindolandaThe photo (source) shows a painted inscription on the neck of an olive oil amphora found at Vindolanda, a small Roman garrison town on Hadrian’s Wall. The most visible part reads Aemiliorum et / Cassiorum, referring to the shippers of the olive oil, the firm or firms of the Aemilii and Cassii.

The only business firms at the time were family ones. Shipping posh olive oil from Provence (1600 km by cart and barge) or Andalusia (ca 2500 km by ship) to an officers’ mess at the cold northern edge of the Roman Empire was a risky business, absent banks, insurance, or bills of exchange. [Update 4/12: I badly underestimated Roman trade finance; see comments.] To spread the risk, it made sense for the Aemilii and Cassii to join forces.

Family firms are popular with the writers of soap operas, in Brazil as much as the USA. (The best British example, the mordant Steptoe and Son about anachronistic rag-and-bone men with a horse-drawn cart, was satire with no pretence at economic realism.) The familiar emotional conflicts within a family are heightened and given wider repercussions by the less familiar but comprehensible context of lots of money. It’s only recently that soaps have ventured into the technically much more difficult environment of the modern bureaucratic corporation (Mad Men, The Office).

Art draws loosely on life, but doesn’t reflect it statistically. It’s said that Victorian novels frequently used both adultery and quicksands as exciting plot devices. From independent evidence, we know that adultery is very common and quicksands are very rare; but you would not learn this from the novels.

In this case the soaps get it more right than the economics textbooks. The vast majority of the world’s capitalist firms are family ones, even leaving out family farms. In the US, 80% of all firms are family-controlled; and “in about 65% of firms with 1993 revenues of over $5 million, at least 50% of the ownership was concentrated in a single family”. And not just small ones: “about 35% of the Fortune 500 firms are largely controlled by family interests”. (Gomez-Meija et al, pdf page 3, paywall; h/t Nadia Tronchoni in El Pais; thanks to Mike O’Hare for forwarding me the paper.)

My daughter Sarah works in one of the many family businesses in the Lille area. The grandest of these are the Mulliez dynasty: the 550 descendants and inlaws of a fertile 1920s patriarch between them control a retail empire guesstimated to be worth €30 bn. It includes 3,051 Auchan supermarkets and hypermarkets (now in Russia and China), hundreds of Leroy-Merlin DIY superstores, and half of the Decathlon sportsgear shops.

You would suppose that family control makes a difference. You would be right. The CW is that family controlled businesses are risk-averse. The paper by Gomez et al has shown that this is only half true.

Their sample was a large and very homogeneous one: 1,237 olive-oil mills in Jaen province of Andalucia, the global centre of gravity of the industry. It’s basically the same business as in 500 BC – grow olive trees, harvest the olives with a stick, crush them in a press, package for shipping in amphora, tin can, glass or PET bottle. The researchers didn’t have records for Roman growers, but they did have them for Spanish ones since 1944, well back into the Franco era. The other thing that makes this a tidy dataset is that the alternative to independence has stayed the same: joining the local co-op, which successive governments have encouraged through tax breaks and a clear legal status.

What Gomez and collaborators found is that Kahneman and Tversky’s prospect theory applies, on condition that the reference point is the continuation of family control. The firms were risk-averse upwards, as the CW proposes, but risk-seeking downwards. They would take considerable risks to survive independently, even when the economically “rational” decision was to join the co-op.

This only makes sense if family control has value independently of the money involved. Gomez et al coin the faintly squishy term “socioemotional wealth”, with less squishy components of identity, independence, and authority.

Let me coin a term myself. In homage to Jean Bodin, the French Renaissance scholar who applied the term to Muscovy and the Ottoman Empire in contrast to seigniorial Europe, let’s call the embedded culture of such family firms patrimonial. The guiding aim is to maintain and if possible grow the patrimony embedded in the firm, and pass it on to family heirs, typically children. It’s closely parallel to the values of the landed aristocracy and gentry in Western Europe. The concept does not extend to all family firms, though. Many family firms – Mom-and-Pop shops say – are too small and ill-capitalised to be patrimonial; their aim may be to allow children to escape the firm into higher-status jobs, rather than take it over.

Why do patrimonial firms survive? If they are risk-averse upwards, and returns are correlated with risk, shouldn’t these firms be supplanted by joint-stock corporations, with their greater capacity to assume and distribute risk in new ventures?

There are several answers. I won’t try to weight them.

  • The principal-agent problem

The joint-stock corporation comes with its own negative baggage of principal-agent problems. Family members are likely to be loyal to the family, pace the soaps. Hired managers may well be normally competent and honest workers. However, others may be crooks (whose aim is to fleece the owners before bailing out), moochers (looking for a comfortable and high-status life at least effort), gamblers (risk-seekers with somebody else’s money), toadies (needy personalities seeking dependence), or bullies (interested in the sadistic exercise of power over lower-ranked employees). These can do much more damage than a simply inadequate family member.

There is nothing new in the principal-agent problem. The Gospels have several parables of unworthy stewards, as a metaphor for mankind’s unsatisfactory relationship to God –  which assumes that the situation and its problems was a commonplace in first-century Palestine. Jesus’ solution was repentance. That of the business schools is stock options and bonuses. These encourage effort, but also short-term optimisation, manipulation of indicators, and cronyism, for example through remuneration committees of closed circles of peers. Fiddling with the incentive mechanism can improve matters a little, but does not make the problem go away.

A large patrimonial firm like the Mulliez operation, which hires many professional managers, can’t avoid principal-agent problems entirely. But it does have two great advantages in reducing them. The first is the intensity and continuity of the supervision the owners can exercise, compared to the sporadic and distant interest of typical shareholders. The second is the possibility of using a network of strategically placed family members to provide a parallel reporting backchannel, rather like the role of Party members in government and business organizations in Communist states. Whatever incentives the Mulliez put in place to align the interests of hired managers with the family’s, they are going to be much harder to game.

The “wisdom of crowds”, embodied in the stock market price, can be discounted. It’s true that it reflects many more different opinions; but it compresses them into a single, horizon-free and very noisy binary signal (share price up = good: share price down = bad). Compare that to the wealth of information available to patrimonial owners. Why, say, are our supermarkets losing market share in white goods in Romania?

The family firm has another winning strategy here: exogamy. If there is no competent son to take over, a daughter can be married to some capable young Dick Whittington. At a pinch, a successor can be adopted, as Julius Caesar did with Octavius. Wealthy bourgeois won’t have the intense snobbery that used to handicap landed aristocrats. The marriage of Prince William to the well-grounded daughter of a self-made businessman and miner’s son, okayed by the Firm after the sobering experience of his father’s disastrous marriage to a neurotic aristocrat, suggests that this may have gone entirely.

  • Innovation born of desperation

Patrimonial firms in really bad trouble become, according to the study, extreme risk-takers. The owners gamble the family wealth to secure the survival of the company. They thus become like venture capitalists. Of course, most such gambles are doomed. But not all. Sometimes the struggling firm comes up with a killer new product. I’m sure this happens; but I’d be surprised if it happens often enough to influence the overall numbers much.

  • Life-cycle business demography

Marx famously came up with the plausible hypothesis of increasing concentration of capitalist firms leading to inevitable monopoly, as in the board game. If you only look at the 100 largest firms in any industry, this makes sense. Over time their ranks thin and they tend to consolidate. But as a long-term historical trend, it doesn’t work. The reason we do not see the large-scale Marxian trend to monopolization is that small firms are being born all the time in great numbers. They die like tadpoles, but enough survive and flourish to keep the distribution roughly constant. These new firms are also, almost entirely, family firms, so the same explanation holds for the survival of patrimonialism.

  • Macro-evolution

In some sectors, the traded joint-stock company actually does dominate: where you need massive amounts of risk capital, combined with economies of scale. Their initial heyday came from railways, and later waves came in steel, cars, electricity, telecoms, chemicals, aircraft, pharmaceuticals, and computer hardware. This looks like a list of the innovative sectors over the last century: but that’s an illusion. Software, music, design, medical care, restaurants, lawyering and machine tools have also done well and innovatively, and don’t need the enormous scale where patrimonialism becomes infeasible. There is also no law that over time the capital-intensive sectors grow faster. On the contrary, the faster growth of services in recent decades has been a lifeline for family firms.

The problem child is finance. There are still a few boutique private firms like Marley & Scrooge (for instance in private equity five out of the top ten are themselves unquoted). Still, insurance companies and banks are almost entirely quoted public companies. Their responsibility for the global financial crisis does not lend much support to the doctrine of the historic superiority of this corporate form. It looks more like a dangerous anomaly.


Two parting thoughts.

The deeper problem with patrimonial firms has nothing to do with the fictitious superiority of the upstart joint-stock corporation. It’s that they embody an incredibly old-fashioned, almost Roman, concept of the family itself. The model is patriarchy – even when the boss is a matriarch. The Industriels du Nord are usually conservative Catholics, with a social conscience underpinned by Catholic social doctrine. (I’ve shopped in Leroy-Merlin stores in France, Spain, and Brazil, and always been impressed with the courtesy of the staff: a value that very probably comes from the top.) There’s a tension with the modern concept of marriage as an affective partnership of equals, and with childrearing for moral autonomy rather than filial obligation. Somebody should write a book on how the Mulliez manage this.

Second, the model supports inequality by design not accident. The failure of General Motors to support the socialised medicine enjoyed by the workers of Volkswagen and Toyota has always looked the triumph of cultural solidarity over economic calculation. But for a patrimonial firm, offloading the cost of medical insurance would also mean a substantial loss of authority over employees – one of the core components of its socioemotional wealth. Patriarchs need the employment tie to be a feudal exchange of loyalty for protection. Similarly, they will generally be conservative in politics. The threat from left-wing politics is far more through death duties than income taxes.

Amphorae at Bodrum castle in Turkey
Amphorae at Bodrum castle in Turkey

Title reference is to the famous lines in T.S. Eliot’s The Waste Land:

Phlebas the Phoenician, a fortnight dead,
Forgot the cry of gulls, and the deep sea swell
And the profit and loss.

Author: James Wimberley

James Wimberley (b. 1946, an Englishman raised in the Channel Islands. three adult children) is a former career international bureaucrat with the Council of Europe in Strasbourg. His main achievements there were the Lisbon Convention on recognition of qualifications and the Kosovo law on school education. He retired in 2006 to a little white house in Andalucia, His first wife Patricia Morris died in 2009 after a long illness. He remarried in 2011. to the former Brazilian TV actress Lu Mendonça. The cat overlords are now three. I suppose I've been invited to join real scholars on the list because my skills, acquired in a decade of technical assistance work in eastern Europe, include being able to ask faux-naïf questions like the exotic Persians and Chinese of eighteenth-century philosophical fiction. So I'm quite comfortable in the role of country-cousin blogger with a European perspective. The other specialised skill I learnt was making toasts with a moral in the course of drunken Caucasian banquets. I'm open to expenses-paid offers to retell Noah the great Armenian and Columbus, the orange, and university reform in Georgia. James Wimberley's occasional publications on the web

29 thoughts on “From Phlebas to IKEA”

  1. Nice post.

    Minor points: IIRC, Roman law did not recognise corporations or any other form of distinction between the owner and the risk – all partnerships were joint and several. The Aemilii and Cassii are unlikely to have shipped their product to Britain on their own account – they would have sold to a merchant, who might then sell to another merchant, and so on. The chain might be of long standing, but loss in any part is to some degree insulated (the same model holds today in the drug trade). But the Romans did have letters of credit, bank drafts and insurance – all these predate coin by millenia. And modern studies suggest the landed gentry were fairly innovative and often very enterprising. They did not end up owning 90% of British real estate by accident.

    1. “They did not end up owning 90% of British real estate by accident.”

      I thought (seriously, not snark) they landed up owning this land because William the Conqueror (and descendants) gave it to them, not because everyone started off equal in some imaginary Arcadia, and they proved more industrious and able to buy off the land from their neighbors…

      (And is “own” the correct term? Isn’t there some strange technical sense in which, even today, the Queen owns all the land and everyone else just has some sort of usage rights to it?)

      1. Medieval law recognised broadly (innumerable variations) two types of land ownership: allodial (more or less perpetual freehold) and usufruct (use-right). We think of feudal as involving usufruct (lord or king owns land, holder has use-right conditional on service). William maintained that all land in England was usufruct – held conditionally. But both law and attitudes continually blurred the distinction: whatever the legal theory, it was and is very hard to take land away from the actual holder without a specific process and cause. Common law recognises this – see the Mabo case in Australia (where the High Court said that under common law traditional aboriginal ownership continued unless specifically extinguished). In England, the Crown has what in the US would be called eminent domain, but this is not the same as ownership. And who had what rights under what conditions changed a lot even under what we lump together as feudal times.

        Anyway, from early C18 on, the English gentry had a combination of management and legal forms that allowed family estates to ride out most kinds of economic or familial downturns, which meant that they steadily accumulated more and more land. Even the great agricultural depression of the 1870s on did not much more than dent their hold, and ownership in England remains very concentrated.

        James – in a class-conscious society, everyone is a snob, pretty much, and class anxiety is pervasive (still is, in my experience of England). But the forms change – C18 England was probably more open to rapid social climbing than, say C19 (David Cannadine is good on this).

        1. Peter

          I remember learning that the great English contribution to law is the Law of Trust. Ie the ability of the aristocratic family to preserve its estate from the depredations of spendthrift future generations, by putting it all into Trust.

          As the book ‘who owns Britain?’ documents, the concentration of land holdings in the UK is so high that it’s embarassing, and efforts are made to disguise it via trusts, offshore trusts etc. (estate duties have a hand in that).

          This helps to explain opposition to wind farms and now solar farms (Griff Rhys-Jones prominent in getting one stopped in Devon recently). The benefits accrue to large land owners, and the visual costs to the communities where they are sited.

    2. Thanks for the correction on Roman trade. But if the Aemilii and Cassii were insulated from the distance risk, why were they in business together? Could be brand recognition. As I wrote, the stuff must have been premium grade: the transport costs greatly outweighed the ex-mill price, so there would have been little point in selling the cheap stuff.
      I never said the gentry weren’t innovative – see also France before the Revolution. Surely it’s not controversial that they were snobs, all over Europe. Look at the class anxiety reflected in Jane Austen’s novels.

      1. The Aemilii and Cassii were probably in business together because they lived in close geographical proximity and almost certainly dominated the local area and controlled its crop of oil. It’s quite plausible that they would have arranged marriages between their children to ensure that the businesses remained in the hands of their descendants. The Cassii pop up 100 years later, still in business, which might suggest that they won out over the Aemilii:

        Also, it’s worth bearing in mind that the Romans did have maritime loans (pecunia traiectitia) that functioned in many ways as a form of maritime insurance. This was, in some ways, an inheritance from the Greek maritime loans (for which we have evidence from Athens, including the first recorded case of an attempted insurance fraud).

        Demosthenes 32 (“Against Zenothemis”) is worth reading. Sections 4 and 5 sum up the alleged situation.

        1. The note on the inscription suggests the oil came from Spain not Provence. The much lower cost of water transport (still true) won out over the shorter distance overland.

          1. James

            re overland v. by sea

            Unless you had access to canals and internal waterways, then the cost factor would have been something like 10x per unit distance, or even greater than that: I don’t know of any estimates, but I could believe 50x.

            Although Rome built exceptional roads, they were built for marching and for couriers, *not* for horse drawn carts– the grades were too steep (a consequence of their straightness, which is notable today in modern England– an old Roman Road is *straight* compared to its medieval or later comperes). There were secondary roads for that. Even in Roman times it was far more efficient to go by sail.

  2. That was a fascinating post.

    The family firm has another winning strategy here: exogamy. If there is no competent son to take over, a daughter can be married to some capable young Dick Whittington. At a pinch, a successor can be adopted, as Julius Caesar did with Octavius. Wealthy bourgeois won’t have the intense snobbery that used to handicap landed aristocrats. The marriage of Prince William to the well-grounded daughter of a self-made businessman and miner’s son, okayed by the Firm after the sobering experience of his father’s disastrous marriage to a neurotic aristocrat, suggests that this may have gone entirely.

    There’s something similar with Japanese family firms. They’ll just unabashedly adopt certain adult people into the family to ensure the survival of the family’s interests (I’m assuming marriage to a family member was or is often involved, but not sure). It appears to make them more successful over time – they outperform family firms that don’t do this in Japan.

    In some sectors, the traded joint-stock company actually does dominate: where you need massive amounts of risk capital, combined with economies of scale. Their initial heyday came from railways, and later waves came in steel, cars, electricity, telecoms, chemicals, aircraft, pharmaceuticals, and computer hardware. This looks like a list of the innovative sectors over the last century: but that’s an illusion.

    In the US, that came from our financial system in the 19th century. The US had a heavily fragmented system of banks chartered on a state-by-state basis, with restrictions on them opening branches in other states. This meant they usually didn’t have the funds to finance large projects on that scale, so US financiers (including some of the most famous ones, like JP Morgan) got the funding for them by consolidating said firms into large enough companies that they could use stock offerings to finance themselves.

    1. Thanks for the info on Japan.
      On the US: when did banks, anywhere, finance long-term capital investments except as go-betweens? Banks by definition borrow short and lend longer. But their stability depends on not going too long. They do hold long-term bonds and mortgages on their balance sheets, because these can always be sold on the securities markets.

      1. Looking at banks (at least 19th century banks) as go-betweens is not really accurate. That’s the deposit/lending part of the business, but the other side of the business was underwriting, where a bank would take debt paper from an enterprise and sell it to other banks, plungers and similar fools. If the paper turned out bad, the bank might well be in trouble, either directly or as a result of deteriorating reputation. So they were intermediaries, but also needed the capital (financial and social) to cover floats during the period when debt issues were in process. Now that float is measured in hours or days at most, then in months.

      2. If I remember, the antebellum Louisiana banking system allowed for long investments up to capital, and demanded short investments to cover its short liabilities. (Capital ratios were much higher in those days.) The Louisiana banking system was maybe the best in the antebellum era–at least, it survived the crash of 1857 unscathed.

        But I think that this refines James’ point, rather than refutes it.

  3. Sarah writes:

    Not a book about the attitude but a dissertation written by a philosophy student re working in Decathlon.

    We also have a “SBRAM” policy (which is attributed originally to either Mulliez or Leclercq [a lesser dynasty, co-owners of Decathlon]): Sourire, Bonjour, Regard, Au revoir, Merci. Even in Head Office you HAVE to say hello to everyone and make an effort to smile.

    The psychological engineering here is quite powerful. Faking a smile improves the smiler’s mood as well presumably as that of the smilee. Eye contact looks strong too. Compare the feudal ceremonial of commendation: the vassal knelt before his lord, who then took his hands and raised him to his feet.

  4. It seems to me that only the form of the family is changing, whereas for many people, “family” still means the “old-fashioned” virtues of loyalty and devotion. It may not be nuclear, it may not be straight, it may not be patriarchal* but I don’t think family is going anywhere, frankly. It is just morphing a bit.

    Also, I do not think it is fair to “blame” employees for all of the principal/agent issue. The other side is, if you’re not family, it may not matter how devoted you actually are. You will always be suspect. (Unless adopted I guess.) Been there. I would hazard a guess in fact that small businesses may often be less pleasant places to work, but, who knows.

    *And it may not really have been before, either. I recall reading one of those company histories in, iirc, a Carls Jr. It may even have been titled, “one man and his dream.” Whereas if you read closely, it turned out, Wifey did at least half the work … every step of the way. Along with having a bunch of kiddies (those Catholics, you know). But I guess “patriarchal” could just mean who got the credit for things, and that problem is still very much with us.

    1. Para 2: how right you are. It annoys me when authors as eminent as Stiglitz analyse employment asymmetries only from the employer’s point of view, and ignore the employee’s risks of signing up with a bad boss, from workplace abuse to tanking the entire company through incompetence. I’ve committed the same offence in this post – but I was explicitly trying to compare two sorts of capitalist. Off the top of my hat, I’m not sure if you can generalise that it’s better to work for a patrimonial or joint-stock firm. The Mulliez social engineering may be a bit creepy, but on the other hand Auchan employees know for certain that they do not work for scam artists but long-term empire-builders.

      1. Good points, it probably doesn’t pay to generalize. Plus, maybe a 2nd or 3rd generation might still have the p/maternalism, but maybe less identified with the company since they didn’t themselves found it. So maybe you could end up with the best of both worlds.

        And really, I suppose it’s the direct boss who matters most to worker happiness.

        1. The Cossacks work for the Czar. Distant owners don’t micromanage shopfloor supervisors, but their policies at the very least set the tone, and it trickles down; in many cases they have strong, firm-wide policies. Who would you rather work for as a cashier – the Mulliez or the Waltons?

      2. My anecdatum is that given a choice, I will never again work for a publicly traded stock company. The implications of agency problems will always kick in eventually. Family operations can be good or bad, but publicly traded companies will all be bad eventually, even if they aren’t now. Watch out for the appearance of outside consultants: a sure sign it is time to update the resume.

        That being said, I would be very interested in an exploration of how family owned companies tend to do once the third generation or so takes over. Some families have an internal culture of long term stable competence, but there also are the families where the third generation was raised in wealth and privilege and eventually assume control regardless of personal ability: the born on third base, rather than hitting a triple problem.

        1. The recent history of the Sea Island Company (St. Simons Island, GA) is a nearly perfect example of “shirtsleeves to shirtsleeves in three generations” for a family company. Small, yes, and marginal compared to many businesses, but still worth several hundred million dollars and the largest private employer in the area. It foundered when the guy with the “III” after his name took over.

  5. There are families and families. I think that James is mostly correct when “family” is an expansive concept. But I’m not sure about nuclear families. My experience is largely with Jewish family-owned firms in Manchester, England. (My relatives!) A typical pattern involves a founding member with several children. The aspiration is that the bright children go to the professions, and the less-bright children (if not affirmatively stupid) take the business. Since Jewish families tend to be small, ownership doesn’t spread much within the family. Often enough, the next generation has nobody who wants to/is able to take the business.

    If the business is doing okay at this stage, it is then sold to the Mitt Romneys of the world. This is the socially useful role of private equity–solving succession problems in family businesses that have not solved the succession problem within the family.

    1. Exogamy is much more difficult as a succession strategy for Jews. I wonder if adoption is also more difficult because of the strength of the Jewish family, making for a shortage of capable Jewish orphans. But fertility among the Orthodox should be a plus. Is there a Lubavitcher parallel to the Mulliez mega-family?

      1. I don’t think that exogamy would all that difficult–there are plenty of entrepreneurial nice Jewish boys around. Except the daughters mess this up by marrying professionals themselves.

        1. Tom Stoppard should write a play about two star-crossed merchants in Verona whose delicate dynastic plans are wrecked by the raging hormones of ungrateful children.

    2. My impression of Mitt Romney’s model was basically: bribe management to borrow company money for Mitt to take control; Mitt then takes huge “consulting” fees for moving operations overseas and/or firing employees, making debt-ridden company look good on paper; Mitt then sells company; company goes bankrupt, and pension fund is destroyed. In short: conspire with management to steal from stock-holders, employees, and retirees. My understanding is incomplete, however, as I think he also managed to profitably destroy family-owned companies as well.

Comments are closed.