Lithograph c1905-1910 showing a wealthy German shop owner (talking to his Jewish neighbour) while a British worker complains about the goods being unaffordable - in the background the unemployed march

Free-trade illusions

Mike Macnair continues his argument against tailing the liberals

In my November 22 article, ‘Free-trade tailism’, I argued that ‘citation grazing’ in Marx and Engels on the question of free trade could not provide the basis for a usable policy on the issue, even if the Marx and Engels quotations were not mutilated in the way the Alliance for Workers’ Liberty did.

There were two reasons for this. The first is that Marx’s actual argument in 1848 for supporting free trade was that it would impoverish the workers and, hence, cause radicalisation and bring on the revolution. This was a line of argument which Marx had already abandoned by the 1860s, though Engels echoed it in 1888 (albeit also giving other grounds). We should know from the experience of the 20th century that Marx was right to abandon the argument. Acute impoverishment of the working class does not produce a strong workers’ movement, but a weak one.

The second reason is that Engels’ 1888 arguments, so far as they depart from Marx in 1848, postulate that protectionism is justifiable in support of what have come to be called ‘infant industries’, but that, when industry is ‘full-grown’, free trade - the normal condition of capitalism - is preferable.

This article will address the historical and (speculatively) the political-economic issues posed. A third article will return, in this context, to the question of what should be the policy of the working class movement on the issue.

Engels’ view that free trade is normal to capitalism reflected the method of Marx’s Capital, in that Britain (‘England’) was taken as the exemplary instance of capitalism. Marx had stated:

In this work I have to examine the capitalist mode of production, and the conditions of production and exchange corresponding to that mode. Up to the present time, their classic ground is England. That is the reason why England is used as the chief illustration in the development of my theoretical ideas.1

The problem with Engels’ argument is that at the very time when he was writing, the main trend on the issue was a movement towards protectionism, imperialism and trade blocs - including in British politics, though the Liberals in Britain managed to resist it for a while.2 This trend continued, until it reached a high point in the 1930s-40s. What then replaced it - the 1947 General Agreement on Tariffs and Trade (GATT I) - was a regime of managed trade, which authorised substantial protection, subsidy and planning regimes. The ‘neoliberalism’ which emerged into ascendancy from the late 1970s on is so-called because it is a return to ‘Manchester liberal’ arguments of the late 19th century, which had been marginal for around 50 years.3

Karl Kautsky argued in the 1890s that protectionism in Germany reflected the continued influence of the landlord class. This argument was hardly plausible in the light of the rise of protectionism in the US from the 1860s, as northern US industrial capital took control away from the southern ‘slavocracy’. It was also problematic even for Germany, where heavy industry and shipping interests lobbied for imperialism and the junker (landlord) Otto von Bismarck for some time resisted these pressures.

Other authors argued that industrial decline had produced a British turn to financialisation and territorial acquisitiveness, which drove rivals to protectionism. What emerged from these debates was the standard ‘Leninist’ argument, that the rise of protectionism, financialisation and imperialism-colonialism resulted from the general decline of capitalism: Imperialism, the highest stage of capitalism.4


A central problem with this is that it failed to account for the fact that imperialism, in the sense the Second International writers and Lenin described - efforts to hold foreign territories in subordination for economic benefits beyond tribute, land-hunger or slave-taking, including for foreign direct investment in production integrated with and subordinated to metropolitan-country production - began well before the late 19th century: it went back to late medieval Venice and Genoa - and in Britain continued through the period in the 1850s-70s, in which the Manchester-liberal ideology, including a degree of hostility to imperialism, was ascendant.5

Theorists of imperialism as an epiphenomenon of capitalist decline have tended to separate late 19th century imperialism from the imperialism of late medieval Venice and Genoa and the early modern Netherlands and Britain - either forgetting the latter altogether or artificially interpreting it as less connected to capitalism than it actually was.

In relation to free trade, the standard narrative is rather different. Early modern ‘mercantilism’ is the tag for the belief that states needed various means, including tariffs and other regulations, to pursue a positive balance of trade and of payments. It is commonly taken to have grown directly out of the economic ideas of the later middle ages; to have been first criticised by the 18th-century French ‘Physiocrats’ and then by Adam Smith; and the truth of free trade is said to have been recognised in the mid-19th century following the arguments of David Ricardo.

Here the problem is that ‘free trade’ arguments go substantially further back. Hugo Grotius’s 1609 Mare liberum (‘the sea is free’), though it presents itself as being about the law of the sea, is actually about the Netherlands claims (against the Spanish and Portuguese) to freedom of maritime trade. The free-trade issue was live in Netherlands debates throughout the period, and the Dutch were imagined as free-traders well before the debates of the 18th century.6 They were not unique: 17th century Livorno was a ‘free port’ and used as an example of free trade.7

Britain at this period moved into protectionism, with the adoption of the Navigation Acts promoting the shipping industry. It remained protectionist throughout the main period of British industrial growth, abandoning the Navigation Acts and the Corn Laws and turning to free-tradism only in the mid-19th century. The process was slower and more incomplete than is sometimes thought, with distinct ‘mercantilist’ aspects remaining in English policy well after the 1849 end of the Navigation Acts.8 The ascendancy of free-tradism was, therefore, brief as well as incomplete.


The other matter appearing from the history is that free-trade regimes are commonly associated with the ascendancy of one nation-state over another. John Selden’s Mare Clausum (‘the sea is enclosed’) published in 1635, is primarily a defence of the English fishing industry against Dutch factory-fleets.9 In the 1639 battle of the Downs the Dutch fleet attacked and destroyed a Spanish fleet very close to the English coast, with the English navy unable to intervene effectively because of its weakness.10 The first English Navigation Act was introduced in 1651, and was similarly primarily directed against the Dutch, who at this time dominated the shipping industry.11

The United States, though politically independent, remained economically dependent on Britain through the first half of the 19th century, through the commitment of the slaveholder producers of agricultural commodities to allying with British lobbyists in order to defend free trade.12 The issue remained politically ‘live’ through the late 19th century even after the Republicans introduced higher tariffs under Lincoln, with British interests supporting free trade supporters in US elections (and suspected by their opponents of more nefarious activities).13

More generally, Ha-Joon Chang’s 2008 book Bad Samaritans points (among other things) to a number of instances where free trade was forced on weaker countries by ‘unequal treaties’ - most notoriously the treaties of Nanking (1842) and Tianjin (1858), and the Convention of Beijing (1860), following Chinese defeats in the ‘Opium Wars’, which among other provisions legalised British sales of opium to China and the shipping of Chinese ‘indentured servants’ abroad.14 There are a good many other examples.

A ‘free-trader’ point of view would require us to suppose that the British government’s military operations in support of British drug-dealers and people-traffickers in China (taking just this example) was an essentially altruistic operation to bring the benefits of free trade to the Chinese people. It is more plausible that the imposition of free trade in this context was a means of subordinating the Chinese for the benefit of British trade and shipping.

This geopolitical aspect of the issue persists in modern times: the USA is not so militarily dominant that it can simply impose ‘unequal treaties’ unilaterally on any other state, though it comes close. Instead we have ‘multilateral negotiations’, which somehow seem usually to issue in results favourable to US (and British) financial capital (though Trump has moved more into bilateral strong-arming). Joseph M Grieco’s study of GATT negotiations in the 1980s around ‘non-tariff barriers’ to trade (ie, regulations which discriminate against foreign suppliers) reached the conclusion that the action of states in the negotiations was primarily motivated by defensive concerns that the loss of regulatory control might also lead to the loss of strategic or military autonomy, and hence to the wider loss of sovereignty in a world which remains one of ‘international anarchy’.15

The consequence is that the issue of protectionism is not only posed in relation to ‘infant’ industries, but also in relation to declining industries (which are, however, of strategic significance).

The clearest example is British and west European agriculture. This was essentially rendered unprofitable in the late 19th century by competition from grain production of the American mid-west and Ukraine/south Russia and (after the invention of refrigerated shipping) from the meat production of Argentina, New Zealand and so on.16 In 1914-18 and in 1939-45 the result was near starvation, as the British blockaded Germany (and in 1940-45 occupied Europe), and high-risk ‘battles of the Atlantic’, as the Germans attempted to cut British supply lines. The result is that domestic agriculture was subsidised from the beginning of World War II - and the Brexiteers, while denouncing the EU’s Common Agricultural Policy, propose to replace it with other forms of subsidy, not to go over to ‘free trade’ in agriculture. States have to be concerned about war risks like these in their peacetime policy.

Political economy

These historical symptoms would be remarkably difficult to explain if anything like Ricardo’s theory of comparative advantage (that efficiency, in the sense of maximising total social welfare, is improved by free trade), or the modern marginalist reinterpretations of it, were true. If so, it should be obvious from experiment that free trade is to be preferred. The same applies if it were the case that, as Engels suggested, free trade is the natural condition of capitalism. There would then be a secular, long-term tendency towards free trade, which would relatively marginalise the counter-tendencies towards protectionism - even if, as Engels also argued, the result was a tendency towards Zusammenbruch, or the breakdown of capitalism as such.

In practice, there turn out to be problems with free trade, which drive not towards the Zusammenbruch, but towards renewed protectionism, and towards much more complicated trade policies, mixing tariffs, non-tariff barriers (perhaps, as in the US, characterised as ‘sanctions’ or as state security measures) and forms of financial manipulation at the expense of ‘competitor countries’.

What follows is necessarily speculative, and hence largely unreferenced.

The starting point is that both classical political economy and marginalist (or ‘neo-classical’) theories, which assert that free markets tend to equilibrium, are straightforwardly false. The nearer any market gets to a free market with free flows of information, the more it tends towards violent instability and swings - evident in financial markets and commodity futures markets. The claim that the activities of arbitrageurs have the effect of moderating these markets towards true values in the long term has been shown to be false.17

If all markets were like financial and futures markets, the price of a bus or train ticket to commute to work, or of a cup of coffee, would vary so violently from day to day or from hour to hour that no-one could plan for everyday activities without massive expenditures of time and resource on financial hedging, and so on. The result would almost certainly be the cessation of many activities that depend on relative predictability.

Theoretically, ‘dynamic stochastic general equilibrium’ models depend on presuppositions which must be false (for example, that there is no biological floor on wages, so that wages can fall without limit until full employment) and more directly relevant to the present are presuppositions which may be false - for example, that there are declining returns to scale. If there are increasing returns to scale, or merely returns to scale are indeterminate, the mathematical modelling does not produce a unique efficiency-maximising equilibrium.

There is not an actual tendency to equilibrium; but in the upward phase of the business cycle, and equally of ‘long cycles’, the appearance of a virtuous tendency to equilibrium growth. The underlying basis of this is that there is actual growth of productive output in a ‘lead sector’ - for example, shipping in the 18th and early 19th century, railways in the mid-19th to early 20th, road transport and air in the later 20th - which drags other sectors behind it.

However, this would still produce violently ‘noisy’ markets, like financial ones, but for two other elements. The first is that there is a large part of the population of economic actors which either cannot act as ‘rational utility maximisers’ (because they work for wages providing only subsistence and a little extra for recreation, holidays and so on); or merely do not want to act as ‘rational utility maximisers’ (because they are not sociopaths). This circumstance produces a substantial stabilisation of ‘real markets’, as opposed to financial markets.

This, however, does not alter the fact that the ‘rational utility maximisers’ may well be unwilling to pay for activities which are unprofitable or return a below-average rate of profit, but are nonetheless essential for continued production and the continued operation of their own activities. Since money inherently tends, in merely random transactions, to accumulate in few hands,18 and is prima facie likely to concentrate in the hands of financial market actors, the result - other things being equal - would be the material breakdown of production (through, for instance, collapsing bridges, and so on).


The second element is that very substantial components of the necessary infrastructure both of production and of markets themselves are provided by non-market institutions. Charities can be left on one side, since their role in providing, for example, roads and bridges - important well into the early modern period - has been marginalised in modern times. But states provide both large amounts of physical infrastructure and market infrastructure, in the form of routine legal enforcement of property rights, debts and contracts. It is the routine state enforcement of debts which allows debt instruments - bills of exchange, promissory notes and so on (and hence banknotes) - and bank account transfers (and hence payments by plastic) to function at all.

There can be no capitalism without forms of credit money. There was already insufficient gold, silver and copper in circulation in late medieval Europe for the transaction needs of the economy of the time without credit means of payment. The case of developed capitalism is a fortiori.

But there is no universal state. Rather, there are competing geographical states. And money is by its nature portable and ‘has no earmark’. Hence, a state has to have means of coercing debtors not to take their money assets and flee the territory into that of another state, if they are to be able to provide debt enforcement and credit money as a service to markets, and must be able to discriminate against actors who are not loyal to the particular state, at least in the sense of paying debts and not engaging in avoidance devices.19

All capitalist states, therefore, must be to some extent ‘mercantilist’: that is, they must if they can discriminate against capitals which are non-national. But, for some states, free trade can be an effective mercantilist policy in the interest of their own (dominant) capitals. Britain in the mid-19th century and the US in the late 20th provide examples.

It is at this point that the indeterminacy of returns to scale re-enters the picture. Assuming that such returns to scale can be neutral or increase, it follows that scale can itself be a barrier to entry into a market, without further protection. That point was made by the EU Commission Competition Committee in 2005, rejecting an alternative (American academic) view that scale can never be an entry barrier.20 Hence, mid-19th century free trade could be a mercantilist policy in the interests of the (dominant) British shipping industry (as it was for the Netherlands in the 17th century). The case is made stronger because the dominant global financial sector was the British, organised at its core around this shipping industry. Free trade need not have been beneficial for all British capitals - it was certainly very disadvantageous to agricultural capital - to be overall beneficial to the British state by serving its dominant sector.

I said that the appearance of equilibrium is produced in economic boom or upswing periods - and that it is so because of the limits on markets. In downswing periods, the irrationalities of markets come into obvious public view - as we have seen, most recently, in the crash of 2008 and its aftermath.

From this point of view, we might be able to suggest tentatively a reason for movements between the ascendancy of free trade as an ideology, on the one hand, and nationalist and protectionist policies as - also - a form of ideology. The boom or upswing period makes it appear that pure market forces produce equilibrium growth without assistance. This appearance wins support for liberalism as an ideology. But, when implemented in practice, liberalism produces market instability - and with it insecurity for the general public - and problems for the state in resourcing its necessary activities. The result is a swing against liberalism. Liberalism will not recover until enough time has passed for its disasters to be forgotten - and until a new prolonged economic upswing has given it new plausibility.



1. www.marxists.org/archive/marx/works/1867-c1/p1.htm.

2. A Howe Free trade and Liberal England 1846-1946 Oxford 1997; F Trentman Free trade nation: commerce, consumption and civil society in modern Britain Oxford 2008.

3. The narrative can be found in outline in BE Moon Dilemmas of international trade Boulder, CO 2000; and in H-J Chang Bad Samaritans: the myth of free trade and the secret history of capitalism London 2008. (I do not mean to endorse the arguments of either book.)

4. For the argument here and in the next paragraph see my introduction to B Lewis and M Zurowski (translators) Karl Kautsky on colonialism London 2013 and references there.

5. See above, note 3.

6. P Brandon, ‘Marxism and the “Dutch miracle”: the Dutch republic and the transition debate’ Historical Materialism Vol 19, 2011, pp106-46; A Weststeijn, ‘Dutch Brazil and the making of free trade ideology’ in M van Groesen (ed) The legacy of Dutch Brazil Cambridge 2014, pp187-204.

7. C Tazzara, ‘Managing free trade in early modern Europe: institutions, information and the free port of Livorno’ Journal of Modern History Vol 86, 2014, pp493-529.

8. The point that Britain remained protectionist through the main period of British industrial development is repeatedly made in H-J Chang Bad Samaritans (see note 2). On the later years of the Navigation Acts and their repeal, see JH Clapham, ‘The last years of the Navigation Acts’ English Historical Review Vol 25, 1910, pp480-501. For the persistence of forms of protection, see JVC Nye, War, wine and taxes: the political economy of Anglo-French trade, 1689-1900 Princeton, NJ 2007 (Nye is a neoliberal equilibrium theorist, but that is no reason to reject the historical data of the book).

9. The book asserted British sovereignty over, at least, the North Sea. On the connection to fishing see M Brito Vieira, ‘Mare liberum vs Mare clausum: Grotius, Freitas and Selden’s debate on dominion over the seas’ Journal of the History of Ideas Vol 64, 2003, pp361-77. It had later broader influence on theories of British imperial claims (see M Somos, ‘Selden’s Mare Clausum: the secularisation of international law and the rise of soft imperialism’ Journal of the History of International Law Vol 14, 2012, pp287-330).

10. There is a convenient short note at www.english-heritage.org.uk/visit/places/walmer-castle-and-gardens/history/maritime-decline-battle-of-the-downs.

11. See the summary at https://en.wikipedia.org/wiki/Navigation_Acts#Navigation_Act_1651.

12. AG Hopkins American empire Princeton NJ 2018, chapter 4.

13. M-W Palen The ‘conspiracy’ of free trade: the Anglo-American struggle over empire and economic globalisation 1846-1896 Cambridge 2016.

14. There is a convenient summary at https://en.wikipedia.org/wiki/Second_Opium_War.

15. JM Grieco Cooperation among nations: Europe, America and non-tariff barriers to trade Ithaca, NY 1990.

16. Grain: see Wikipedia - https://en.wikipedia.org/wiki/Great_Depression_of_British_Agriculture (mainly referring to US grain exports); on Ukraine/Russia, ME Falkus, ‘Russia and the international wheat trade, 1861-1914’ Economica (NS) Vol 33, 1966, pp416-29. Meat: J Rees, ‘The invention of refrigerated transport and the development of the international dressed meat trade’ in K Gavroglu (ed) History of artificial cold, scientific, technological and cultural issues Dordrecht 2013, chapter 13.

17. A Shleifer Inefficient markets: an introduction to behavioral finance Oxford 2000. Shleifer’s assumptions are marginalist: he is merely arguing against the claim that financial markets are efficient. The point that they are not efficient does, however, have large implications for free-market equilibrium claims.

18. I Wright, ‘The social architecture of capitalism’: https://arxiv.org/pdf/cond-mat/0401053.pdf.

19. See, for example, the discussion of early forms, where the foundations become apparent, by Y González De Lara, ‘The secret of Venetian success: a public-order, reputation-based institution’ European Review of Economic History Vol 12, 2008, pp247-85.

20. Directorate for financial and enterprise affairs, competition committee, roundtable on barriers to entry (note by the European Commission, DAF/COMP/WD(2005)59).