Wilson did not pursue the exchange with Isles. His critique of Isles was only a passing broadside. His leading concern was the publication of the contents of his DPhil and PhD: Capital imports and the terms of trade examined in the light of sixty years of Australian borrowing, that was published in 1931 by Melbourne University Press, through the good offices of Giblin and Copland, after Angus and Robertson had declined it. His debt to Viner, Wilson imposingly admits: ‘Where I have disagreed with him I am probably wrong. If I am right the triumph is not mine but my master’s’.

Wilson’s thesis addressed this question: ‘How is a country’s terms of trade affected by the flow of capital in (or out) of it?’ Will a country sending sums abroad find its imports cheaper, or more expensive in terms of its exports? This had become topical in the post-war world, with the prospect of large sums flowing from Germany to the victorious powers. But the question was as old as Mill. And the received answer was also as old as Mill. International borrowing (‘capital imports’ or ‘capital inflow’) improves the borrowing country’s terms of trade: its imports will become cheaper relative to its exports. This is because to borrow from abroad allows payment for part of imports by means of the loans from abroad. The borrowing country is, therefore, relieved of the necessity of exporting a value equal to its imports. Exports are consequently contracted by the home country. And with an unchanged world market for its exports, that reduction will make its exports more valuable. The terms of trade improve.

The logic of Mill’s argument can be extended to other questions. What will happen when the loan has to be repaid and serviced? Any net repayment means that the country must now export a value greater than its imports, in order to pay not just the imports but also the debt service owed to the foreigners. The terms of trade must fall. This conclusion about debt service had an easy application to the consequence of German reparations to the Allied Powers. As servicing debt is like a reparation, reparations will necessitate a depreciation in the German terms of trade, which will amount to an additional burden to be carried by Germany. This phenomenon – reparations deteriorating the terms of trade and making the burden of reparations even worse – was known as the ‘Transfer Problem’, and generated a considerable literature.

Wilson announced in Capital imports and the terms of trade that he would ‘contest the accepted views’ on this issue. There was, he said, ‘no theoretical grounds’ for Mill’s assertion. However, said Wilson, there did exist a strong theoretical expectation concerning a distinct, but related, question that classical theorising had never pondered: What was the impact of capital inflows on the prices of goods that could be neither imported or exported? Wilson argued that the price of such ‘non-tradeables’ relative to other goods would rise with capital inflows. This thesis of his concerning capital inflows and ‘the second terms of trade’ – the price of tradeables relative to non-tradeables – was to become a recurrent theme in Australian theorising throughout the twentieth century, and would bear fruit ultimately in the Gregory Thesis.

Wilson’s basic insight with respect to capital flows and the terms of trade is that any capital inflow amounts to a transfer of purchasing power from the lending economy to the borrowing economy. This implies that, while the increase in purchasing power induces the borrowing country to buy more of its own export goods and therefore supply less of it to world markets, at the same time the lending country will be buying less of the borrowing country’s export good, on account of its reduced purchasing power. There is, therefore, a contraction in the demand for the export good that accompanies the contraction in its supply. It is perfectly possible that the contraction in demand for the good will equal the contraction in supply. In that case the equilibrium value of the export good is unaltered, rather than increase. A capital inflow need not increase the value of the borrowing country’s export goods. Correspondingly, a capital outflow need not reduce the value of the borrowing country’s export goods. Hence Mill was wrong in unconditionally predicting transfers and outflows abroad would worsen the terms of trade. One consequence of this conclusion was that the ‘Transfer Problem’ might be an imaginary terror.

Wilson argues his claims in terms of the beggar Lazarus and Dives the rich man in the parable of the rich man and Lazarus, (Luke 6:19–31) ‘clothed in purple and fine linen, and fared sumptuously every day’. It is then expounded in cumbersome numerical examples. But if Wilson’s theoretical methods were obsolete, his empirical technique was modern. With the methods he had been taught in Chicago, he used Australia’s lavish borrowing as a test case. ‘In this as in other directions Australia has almost turned herself into the social and economic laboratory which is often sought but rarely found’ (Wilson 1931b, p. 3). With data for borrowing and the terms of trade for 1893–1913, Wilson computed the coefficient of correlation between capital imports and the ratio of export to import prices to be -0.29. Borrowing deteriorates the terms of trade (the opposite of what Mill maintained).^[16]

Capital imports and the terms of trade was reviewed in the American Economic Review (‘closely reasoned’), the Journal of Political Economy (‘an important contribution to the literature of international trade’), and by Harrod in the September 1932 issue of the Economic Journal. Judging the book to be ‘extremely interesting and suggestive’, and ‘a valuable contribution to the literature on international trade’, he seized on Wilson’s suggestion that the terms of trade impact of capital movements was ambiguous, and re-presented the case with a greater lucidity than Wilson. Hicks wrote to Harrod to say he was ‘very interested in your review of Wilson’.^[17] More cautiously, D.H. Robertson told Harrod: ‘I haven’t yet coped properly with your review of Mr. Wilson on capital transfer, and am wondering whether you aren’t between you really saying the same thing as Pigou in his highly concentrated and still unpublished paper’.^[18] Indeed, in the following (December 1932) issue of The Economic Journal, the kind of ideas Wilson was exploring were presented by Pigou definitively, if rather laboriously, in mathematical terms.

But the burgeoning literature on capital flows and the terms of trade ignored the second theme in Wilson’s logic: the relations between capital inflows and the ‘second terms of trade’ – the price of tradeables to non-tradeables – a theme that was to have resonance in the 1970s, with notions of the ‘Dutch Disease’ and the ‘Gregory Thesis’.

Wilson’s deliberations on this issue began with the simple distinction between ‘exportables’ and ‘domestic’ goods, or, in equivalent modern language, ‘tradeables’ and ‘non-tradeable’. A cigar is exportable. A haircut is not. It is a non-tradeable. Undeniably, British demand for German non-tradeables (say, German haircuts) must be lower than German demand for German non-tradeables (the German haircut). Therefore, any transfer of purchasing power from Germany to Britain must reduce demand for German non-tradeables, and so reduce their price relative to German and British exportables. By a parallel logic, the transfer of purchasing power to Britain would increase the value of British non-tradeables. Such a transfer of purchasing power would comprehend not only ‘reparations’ but also foreign aid, capital inflow (‘capital imports’), or the discovery of some natural resource that earns large economic rents. In all cases, the relative price of non-tradeables in the country receiving the inflow will increase. Gratifyingly, while Mill’s thesis about capital flows and the terms of trade was not verified empirically, Wilson could report that ‘some verification is found in Australian experience for the proposition that imports of capital tend to be positively correlated with increases in the ratio of “domestic” price level to the price level of “international” commodities’.

It was Australian economists who were to cultivate Wilson’s ideas on the ‘second terms of trade’, especially Trevor Swan (1918–89), who was appointed in 1950 as the foundation professor in economics at the Research School of Economics at the Australian National University (ANU), after Wilson had declined the post. Swan was no protégée of Wilson. In Australia his teachers were R. C. Mills, Ronald Walker, S.J. Butlin, John La Nauze; and overseas Michal Kalecki, James Meade, Nicholas Kaldor, Richard Stone, and Arthur Smithies. Nor had he received Wilson’s patronage: Coombs performed that role for Swan. But in early 1950s Canberra, the new professor was in regular contact with the now senior mandarin of the Treasury, Roland Wilson. It was a source of pride to Swan that he, the professor, regularly lunched with the policy-maker Wilson.^[19]

The first fruit of Swan’s digestion of Wilson’s thesis was the ‘Swan Diagram’ (Swan 1955), which demonstrated how, in a neo-Keynesian context, capital inflows would necessitate an appreciation in the value of services (that is, non-tradeables). Wilfred Salter (1929–63), at Swan’s department in the ANU’s Research School of Social Sciences from 1956 until 1959, recast the same ideas in neoclassical form in the ‘Salter Diagram’ (Salter 1959). Perhaps the subsequent development of Wilson’s ideas closest to Wilson’s original approach was that undertaken by Allan Ross Hall, appointed in the early 1950s to the Research School of Social Sciences, in his paper ‘Capital imports and the composition of investment in a borrowing country’ (in Hall 1963). In this paper Hall wrote that ‘conclusions of Roland Wilson’s Capital imports and the terms of trade … have not been fully appreciated by Australian economic historians (including the present writer a dozen years ago)’.

But the most significant revival came with Trevor Swan’s return to a public stage in the early 1970s, after some ‘difficult years’, with his review of ‘Overseas investment in Australia. Treasury economic paper no. 1’ in the Economic Record (Swan 1972). According to Swan:

Swan subsequently notes in his review:

The key application of the notions that Swan had cultivated was now about to take place, in the ‘Gregory Thesis’. R. G. Gregory had joined the Research School of Social Sciences at the ANU in 1971, having trained in international economics. In his 1976 paper, ‘Some implications of the growth of the mineral sector’, Gregory pressed a simple but socially significant implication of Wilson’s thesis: that an increase in the value of non-tradeables would cause a depression in the production and incomes in the import-competing sector. The paper resonated sharply with current events. Booming values in the prices of fuels during the 1970s and 1980s and the associated inflows to develop them had raised the prospect of de-industrialisation.^[20]