An unobserved model of Byzantine economic development

After reimmersing myself in the literature of frontiers back in summer of 2017, I deduce from the blog stubs I left for myself that I must then have made a proper attempt to read Michael Hendy’s The Byzantine Monetary Economy.1 This is a monster tome which was supposed to be one of three and still contains a vast amount of material whose relevance to the exact topic is hard to see, but which also throws out important points and valuable insights as if they were incidental; it really needed an editor, but the legend goes that Hendy told Cambridge University Press that if they changed a word of it he’d cancel his contract with them, and somehow they wanted the book badly enough that this cowed them. So it went out as he wanted it even though it’s hard to understand, as a reader, why that was. In any case, despite being thirty-plus years old it’s still important and, I guess because I was by now writing up the work that would become my ‘Middle Byzantine Numismatics’, I set out to read it.2

Cover of Michael Hendy's Studies in the Byzantine Monetary Economy c. 300-1450 (Cambridge 1985)

Cover of Michael Hendy’s Studies in the Byzantine Monetary Economy c. 300-1450 (Cambridge 1985)

One of the major controversies in which Hendy repeatedly intervened during his rebarbative scholarly life was that of the importance of commerce to the Byzantine Empire, both economically and ideologically. Most people have been at least ready and in some cases downright keen to see the emperors as wanting to grow their commercial economy, even though they are sometimes hard-placed to explain why. For Hendy, however, this was anything but a given, and he saw the primary purpose of the coinage, for example, as to enable the tax system and the payment of the army, not to facilitate market exchange. Without wanting to spoil the book for you, he in fact went on to argue that the Byzantine Empire resisted commercialisation to the point that this became the reason that the Italian city-states with which they reluctantly dealt were able to out-compete them and drain the empire’s resource westwards.3 I do personally find him persuasive on this general score, I admit, but of course he was publishing under Thatcher and no-one was interested in anti-commercial scholarship as the 1980s boomed and academics were settling into how we justify the Great Divergence without having to give up our global predominance.

Nonetheless, he began, or nearly began, with a stab at economic modelling as applied to past societies that I think bears thinking with even now. It should be said that Hendy was just as prepared as his rivals to build elaborate hypotheses on shaky figures—he spent seventy pages here on reconstructing the Byzantine imperial budget, largely on the basis of eighteenth-century Ottoman figures, for example—but he obviously thought his were better, because he liked to attack others’ anyway. As witness, on p. 7 he has a set at someone who had applied Fisher’s Equation to the debasement of the Byzantine coinage in the eleventh century.4

Graphic description of Fisher's Equation of Monetary Quantity

If you’ve not met Fisher’s Equation, here is a summary representation, linked through to a pretty clear explanation; I would try myself, but this post is already pretty long and in-depth economics will not help…

Because he never wrote anything briefly when you would like him to have, I summarise how Hendy dealt with this rather than quote. Firstly he admitted that we probably do now have a decent grip on how that debasement unfolded, in which, ironically, he was probably wrong.5 He then admitted that in an economy where there was effectively no credit, and therefore no elasticity in the money supply, restricted as it was by available precious-metal, the application of Fisher ought if anything to be simpler than in a modern economy. But because the coinage was not, as he saw it, a commercial instrument and not made in quantities intended for it to be one, and was thus distributed not where trade required it but where soldiers and state operatives spent it; because transport was slow and its costs away from water very high, with consequent limits on what could be traded and how far; because, “the producer was almost invariably the distributor and/or the seller”; and because a really substantial part of the empire’s wealth was owned by the emperor, a few landed magnates and the Church, and thus immobilised…

“In the light of all these circumstances separately or in combination, and despite wide-ranging claims to the contrary, it is at least questionable whether the application of Fisher’s Equation has much, if any, relevance to the situation, and whether the pre-conditions necessary for its operation in any chronologically and geographically uniform, and in any detailed, fashion existed.”6

And you can see from that both why Hendy is little quoted, if much cited, and how his book ran to 773 pages. Even so, there are still bits one wants to quote on themes like this…

“These observations… are intended to suggest that it is on the one hand unacceptable for the numismatist, in accounting for some monetary phenomenon, to connect it with a contemporary ‘economic crisis’ (for the basic distinction between a financial and an economic crisis is one that is scarcely ever made), the existence of which is asserted through reference to another such assertion, which turns out to be based on a statement in George Ostrogorsky’s History of the Byzantine State – however distinguished that author, and however valuable that work. But they are also intended to suggest that it is on the other hand equally dangerous, that is dangerous enough to be unacceptable, for the numismatist, in accounting for some other monetary phenomenon, to insert it into a precise mathematical interrelationship evolved in the light of modern monetary theories and conditions. In general, if in no other sense, the result is thereby lent an entirely spurious air of precision and authority, and the nature and mode of operation of the ancient or mediaeval monetary economy involved is effectively never questioned.”7

You see what I mean by now, I guess. Part of me wants to yell “hurrah” and the rest is saying, “Wait, where was all this going again?” and “Could that maybe have been shorter, with fewer subclauses, or else in more than three sentences?” and unhelpfully unsympathetic things like that. I suppose that the general point here is that a model that is never tested against data or accurately set into context can never be proven or disproven.8 Of course, as I say, that didn’t stop Hendy coming up with his own, and what I want to do with the rest of this post is extend one of them for fun. You see, having got to that bit quoted above where he concluded that Fisher’s Equation wasn’t going to work here, he tries to explain the state’s resort to debasement by other means, for which the chief reason was its inability to extract very much money from its leading aristocrats. (He elsewhere argues that the wealthiest Byzantine magnates could severally possess enough to come close to equalling, in their total worth at least, the entire state budget, and while the comparison relies on the accuracy of his reconstructed budget, the figures for aristocratic wealth, at least, are contemporary ones.9) To their wealth, however, there was little alternative, given the probable insufficiency to make up the gap of what could be got from overtaxing the peasantry—which anyway tended simply to drive them into dependency upon those untouchable aristocrats instead.10 Sorry: once you start trying to think with Hendy it’s apparently difficult not to write like him. I’ll fight it.

This got me thinking, anyway, and what I thought is that it has implications which Hendy did not draw out. The tenth century was a time of recovery for the Byzantine Empire, territorially and militarily speaking, but by the end of it, nonetheless, the state was nearly bankrupt. (That is usually put down to Alexios Komnenos’s loss of Anatolia, but he inherited the financial situation, he didn’t create it.11) This would be exactly that distinction between economic and financial crisis Hendy was griping about, I guess. So, OK, let us suppose, as part of another of these untestable models, that, say, the top 5% of the Empire’s population was effectively immune from serious taxation, but that the rest was not. In that case, wealth that accrues to those possessors was effectively amortised from the state’s resources. If the economy grows in such a way that the aristocrats do well out of it, as it seems to have done in the Byzantine tenth century, the figures might work out in such a way that the population overall got richer but the state still got poorer. Now, obviously, one solution might indeed be to try and boost the commercial side of the economy and make up the difference on tolls and sales tax, but since the big aristocrats were essentially autarkic, or could be, that would not liquefy their wealth back to where the state could siphon it off again. So instead, the solution that probably works best for the state is actually to slow the economy down, to encourage deflation and to generally attack the value of wealth until the status differential between the aristocracy and the state has been restored. In that case, overtaxing would not be a desperate tactic to which a bankrupt government was forced despite the damage it must cause to the productive sector; that damage would actually be the point and overtaxing the whole strategy.

Base-silver trachy of Emperor Alexios I Komnenos struck at Thessaloniki in 1081-1092, Barber Institute of Fine Arts B5532

The expedient to which the state had been reduced: a supposedly silver trachy of Emperor Alexios I Komnenos struck at Thessaloniki in 1081-1092, Birmingham, Barber Institute of Fine Arts, B5532

In all of that case, then, it could be very much in the interests of a state constructed as we’ve just imagined to hurt its own economy, in order to be able to appropriate more of what was left. Perhaps that is in fact what Alexios I was doing when he reformed, causing what must have been great expense and considerable monetary shortage, that duff coinage!12 It’s obviously not a very capitalism-compatible model, but I think it’s where Hendy was pointing. That he didn’t get there may have as much to do with the arrangement of the book—in which, within six pages from here, he was having to say, “It may be thought that I have wandered far from the customary or even proper preserve of the numismatist, in discussing such questions as erosion, predominant forms of land-use, and twelfth- and thirteenth-century frontiers – and so, perhaps, I have…”—as any capitalist sympathies of his own.13 I’m not even sure it matters what he was ideologically, because what concerned him was how this other society had worked. The political climate of the age may be why no-one else picked up this idea, and maybe I would not have spotted it lurking before 2008 either. But what are we doing this study of the past for, if not to find alternate ways for human societies to do things? I’m not saying this one’s an obvious winner—though I often have to remind my students when they write about the inevitability of the Empire’s decline that it lasted more than a millennium, however variable its health in that time, so its ways of managing politics and change might still work out better than ours—but at least it is one of those alternatives that we are now, maybe, able to see and think with.


1. Michael F. Hendy, Studies in the Byzantine Monetary Economy, c. 300-1450 (Cambridge 1985).

2. Jonathan Jarrett, “Middle Byzantine Numismatics in the Light of Franz Füeg’s Corpora of Nomismata” in Numismatic Chronicle Vol. 177 (2017), pp. 514–535, which uses Hendy quite a lot.

3. Hendy, Studies, pp. 221-251 on the economic bases and 554-602 for the trade situation.

4. Ibid., pp. 157-220 for the budgetary reconstruction and pp. 613-618 for a worked-out comparison to the Ottomans, on the basis of the same figures he used to construct the Byzantine budget, a circularity he doesn’t seem to have considered. The person who had misapplied Fisher’s Equation is not named by Hendy, but it’s pretty likely that he was referring to Cécile Morrisson, “La dévaluation de la monnaie byzantine au XIe siècle : essai d’interpretation” in Recherches sur le XIe siècle, Travaux et Mémoires du Centre de recherche d’histoire et civilisation de Byzance 6 (Paris 1976), pp. 3–47, reprinted in eadem, Monnaie et finances à Byzance : Analyses et techniques, Collected Studies 461 (Aldershot 1994), chapter IX, which does indeed apply Fisher to the eleventh-century valuation and which Morrisson was still defending as such an application in eadem, “Money, Coins and the Economy” in Paul Stephenson (ed.), The Byzantine World (London 2012), pp. 34–46 at p. 41 n. 33.

5. Hendy, Studies, p. 3; but Cécile Morrisson, J.-N. Barrandon and Jacques Poirier, “La monnaie d’or byzantine à Constantinople : purification et modes d’altérations (491-1354)” in Morrisson, Claude Brenot, Jean-Pierre Callu, Barrandon, Poirier and R. Halleux (edd.), L’or monnayé I : purification et altérations de Rome à Byzance, Cahiers Ernest Babelon 2 (Paris 1985), pp. 113–187, the same year demonstrated that the debasement had in fact begun at a lower level in the late tenth century and that the eleventh-century tipping point was an illusion presented by the written sources.

6. Hendy, Studies, p. 5.

7. Ibid. p. 7.

8. For me, the archetypal case of this is Keith Hopkins, “Taxes and Trade in the Roman Empire (200 B.C.–A.D. 400)” in Journal of Roman Studies Vol. 70 (London 1980), pp. 101–125, which is so obviously and openly founded on no evidence except the author’s own expressed preconceptions that I don’t understand how it got published, let alone became a standard reference.

9. Hendy, Studies, pp. 201-220.

10. See Peter Frankopan, “Land and Power in the Middle and Late Period” in John F. Haldon (ed.), The Social History of Byzantium (Chichester 2009), pp. 112-142.

11. The politics are best retold in Peter Frankopan, The First Crusade: the call from the East (London 2012), pp. 42-70, but on the finances specifically, see, with care, Cécile Morrisson, “La Logarikè : réforme monétaire et réforme fiscale sous Alexis Ier Comnène” in Travaux et Mémoires du Centre de recherche d’histoire et civilisation de Byzance Vol. 7 (Paris 1979), pp. 419-464, repr. in eadem, Monnaie et finances, chapter VI.

12. The more normal position on this is summarised, with references, by Alex Nobes, “The economic and monetary policy of the Byzantine Empire under Alexios I Komnenos” in Rosetta Vol. 11 (Birmingham 2011), pp. 56–71, online here, good work for an undergraduate journal. However, I disagree with him (and indeed Morrisson, “La Logarikè”, on which he rests here) that Alexios’s coin and tax reforms increased state revenue fourfold; I’ve run those numbers as best I can and I’m pretty sure that they come out meaning that he managed to return the levels of taxation to roughly pre-debasement levels by shifting them onto originally supplementary levies that were now paid in the new coin, rather than the debased valuations of the old core taxes; but the roughly thousand-fold increase in notional tax liability that resulted probably amounted to a slight decrease in overall revenue, that’s how bad things had got. So the reform’s purpose can’t have been just that, or you wouldn’t bother, and I don’t see why it shouldn’t have been deflationary as well as stabilising.

13. Hendy, Studies, p. 13.

7 responses to “An unobserved model of Byzantine economic development

  1. This sounds plausible to me. It raises questions about other ‘expensive’ coin reforms in medieval societies – Alfred the Great and Ceolwulf II in the 870s, for example. Economic benefit and commercial activity were not the only possible motivations.

    I’m also glad not to be the only person who finds Hendy hard going. The Byzantine Monetary Economy book in particular is horrendous. I doubt any publisher would take it nowadays.

    • Thanks for these thoughts, Rory; also, Happy New Year! I hesitate before applying the same logic to Alfred and Ceolwulf II, just because I wonder if we know enough about the tax system they were implementing to be sure how the costs would have worked out, but I guess there is a level at which they just had to get hold of quite a lot of silver to make it work whatever the tax system was. But in their circumstances, I’m pretty sure the ability to raise taxes that the Vikings would accept as tribute—and maybe troops as pay—must have been the overwhelming concern, and in that respect Alexios’s crumbling Anatolian frontier may be a parallel.

      And no, I admire Hendy very much as thinker, but only the oddest of persons could admire his writing style for anything other than extremity…

      • Happy new year to you too! I’m wary of tying the 870s reform too closely to military activity and tribute payments, and I quite agree that they weren’t operating with anything like the same taxation systems in England. But I like the idea that the cost of a coin reform could be passed on to the customer, as it were. It’s very much worth pursuing the comparison, as western European and Byzantine (let alone Islamic) monetary histories still tend to be treated as separate entities for most purposes, and operate under different scholarly rules and traditions.

  2. I swither between the view that MV = PQ is just an empty tautology, and the view that it is a dud beyond all rescue. Maybe it’s both.

    How do you get an independent measure of each of the four variables such that you can test the equation? As far as I can see, you can’t.

    If Fisher was the chap at Yale whose economic opinions in the late twenties proved to be such utter tosh – the Great Crash of 1929 being compelling evidence to that effect – then perhaps its failings should be no surprise.

    • Yes, it does turn out to be that Fisher, although my same investigation revealed that many economists would like to rescue his work from that particular mistake. I am close to your views on how deep the discipline can go given its evidential base, though, so don’t really care whether they’re right or not.

      In a medieval situation, the only values we could reasonably hope to hypothesize in the best of conditions would be M, albeit by using ‘the kind of maths we should not do‘, and P, from the occasional attempts by governments to fix prices—and even then you only have a baseline value which you know automatically was being flouted. By the time we get to full-on state and/or mercantile accounts in the thirteenth or fourteenth century, we might also be able to hypothesize Q from that information; and then, of course, if you believed Fisher you’d be able to deduce V. But I think that’s the only direction the solution of the equation can take, and not for very many societies prior to government-run sales taxes (and the survival of that data—but for that last factor many an Islamic régime of my period would qualify…).

  3. Pingback: Seminar CCLI: I guess sometimes even Michael Hendy was wrong | A Corner of Tenth-Century Europe

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.