The Italian Interregnum and the 1945 Reform: Money under Occupation and Reconstruction

The forgotten founding

The 1945 proclamation that established the modern Ethiopian birr is, in the standard telling, a footnote: a routine act of post-war monetary housekeeping, comparable to the parallel reforms that gave most newly independent African states their own currencies a decade or two later. This article argues that it deserves more attention than it gets, because the institutional architecture set up in 1942–45 contained, in the form of three specific provisions, the mechanism by which every subsequent Ethiopian regime would be able to do what Ethiopian regimes have in fact done: finance their deficits through the central bank.

The argument is institutional, not personal. The 1945 founders were not malign. The provisions they wrote into the State Bank’s charter were standard for central banks of the period and were less permissive than many. The point is that the same institutional features that look benign in the hands of a small, externally constrained imperial government become catastrophic in the hands of a large, externally unconstrained military government, and that the institutional design therefore matters more than the personalities. To see this clearly, you have to look at the founding documents.

But first you have to look at what those founding documents were responding to, which is the Italian monetary occupation of 1936–41.

The lira interlude

The Italian invasion of Ethiopia in October 1935, the fall of Addis Ababa in May 1936, and the proclamation of Italian East Africa (Africa Orientale Italiana) made Ethiopia, formally, a province of an Italian empire that included Eritrea and Italian Somaliland. The monetary consequences were immediate. By the proclamation of 15 July 1936, the Italian lira became the legal currency of Italian East Africa, and Ethiopian banknotes were withdrawn from circulation at a rate of 3 lire per talari.¹ The Maria Theresa thaler, still the dominant coin in rural circulation, was assigned an official exchange rate that the Italian authorities then revised upward repeatedly — to 4.50 lire, then 5.00, then in stages to 13.50 — in a deliberate effort to incentivise the surrender of silver coins to the colonial monetary authority.²

This is, viewed from a distance, a textbook example of what a state currency can do that a commodity currency cannot. Italian authorities could and did manipulate the official lira-thaler exchange rate to extract real resources from the Ethiopian population. The people who held thalers — the rural Ethiopian peasantry, primarily — were being told, through the rising lira-per-thaler rate, that their silver was worth more in lira than it had been the day before, in the hope they would surrender it. Many did. Many others did not. The Italian authorities never succeeded in fully displacing the thaler from the rural economy, in part because the rural economy did not trust paper money issued by an occupying power, and in part because the thaler kept its silver content regardless of what the lira-per-thaler rate said it was worth.

The Italian period also, importantly, gave Ethiopia its first sustained encounter with paper-money inflation. Wartime Italian fiscal pressure produced rising prices across the colony from 1939 onward; by the time British and Ethiopian forces drove the Italians out in 1941, the lira economy was in significant disorder.³ The Ethiopian population — at least the parts of it that had been monetised under the Italian system — had learned, for the first time at scale, what it felt like to hold a currency that lost value while you held it.

The point worth seeing here is that the founders of the 1945 system had this experience fresh in mind. The British administration that liberated Ethiopia and managed the transitional period from 1941 to 1944 ran the country on the East African shilling, issued by the East African Currency Board in Nairobi, with a fixed peg to sterling.⁴ This worked, after a fashion, but it placed Ethiopian monetary policy in foreign hands. The 1942 negotiations between the Ethiopian government and the British administration produced the agreement that Ethiopia would have its own currency again, and the 1945 proclamation gave it one. The motive was, in significant part, to escape the experience of having someone else’s monetary policy.

The 1945 architecture

The State Bank of Ethiopia, established by proclamation in 1942 and given its currency-issuing mandate in 1945, was a hybrid institution: it functioned as both the country’s principal commercial bank and its central bank.⁵ This dual function was not unusual for the period — most newly independent countries used similar arrangements — but it had an institutional consequence that the founders of the system did not fully appreciate. When the central-bank function and the commercial-bank function were exercised by the same institution, the conflict of interest was internal rather than external. The bank could expand credit to itself, in effect, by booking the credit in one ledger and the deposit in another. The discipline that comes from a central bank having to say no to a separate commercial banking system was absent.

The 1945 proclamation set the Ethiopian dollar — later renamed the birr in 1976 — at a gold parity of 0.355468 grams of pure gold per dollar.⁶ This was, by design, the same gold content as the US dollar of the period, which set the cross-rate at parity for a brief period before being revised. The currency was backed, on paper, by reserves held in foreign currency and gold. The proclamation contained provisions limiting the State Bank’s lending to the government — but the limits, as in most central-bank statutes of the era, were expressed as percentages of the previous year’s tax revenue and could be exceeded with cabinet approval.

What is worth noting is the structure of these limits. They were procedural, not absolute. The State Bank could not, by itself, decide to lend to the government beyond a certain ceiling; but the government, by its own decision, could authorise a higher ceiling whenever it needed one. The discipline was, in effect, self-enforced by the government on the government. The arrangement worked through the 1950s and 1960s because the imperial government chose to enforce it; it would fail under the Derg because the Derg chose not to.

This is the founding institutional weakness this article is naming. The 1945 system did not give the central bank independence from the government. It gave the government a central bank it could use, with restraint or without, as it preferred. That preference, in the imperial period, was for restraint. The preference of every subsequent government would be different.

The 1963 split — and what it did not change

The 1963 reform separated the central-bank and commercial-bank functions, creating the National Bank of Ethiopia as a stand-alone central bank and converting the commercial side into the Commercial Bank of Ethiopia (CBE).⁷ This is usually treated, in the standard histories, as a maturation of the system — the creation of a “modern” central-bank structure to replace the hybrid State Bank.

It was that, in part. But the 1963 reform did not change the feature that mattered most. The NBE remained an organ of the imperial government, its governor appointed by and serving at the pleasure of the executive, its mandate broad rather than narrow, its statutory limits on lending to the government framed as ceilings the government could itself raise. The 1963 NBE was a central bank in name. It was not, in any meaningful sense, an independent monetary authority. The lever that the Derg would pull in the late 1970s to finance its wars and famines — the lever of central-bank advances to the Treasury — was present in the 1963 statute as much as in the 1945 one.

And the Commercial Bank of Ethiopia, separated from central-bank functions in 1963, became something else worth flagging: a state-owned commercial bank with a near-monopoly on the banking sector, dependent on government patronage for its operations and increasingly used as an instrument of state policy. The CBE has been, ever since, simultaneously a normal commercial bank serving private customers and a parastatal lender of last resort to whichever state-favoured industrial projects the government wanted to support. This dual identity — bank and policy instrument — would matter enormously after 1991, when the EPRDF’s “developmental state” architecture used the CBE as one of its principal tools.

The institutional inheritance

What did the 1974 revolution inherit, then?

It inherited a central bank (the NBE) with statutory provisions allowing it to lend to the government, with limits the government could itself raise, and a governance structure that placed it firmly within the executive branch. It inherited a state-owned commercial bank (the CBE) with most of the country’s deposits and most of the country’s loan book, available as a policy instrument. It inherited a fixed exchange rate (2.07 birr/USD, established in 1949 after Bretton Woods stabilisation) that the NBE was committed to defending with reserves it did not fully control.⁸ It inherited an IMF membership and the standard suite of access rights that came with it. And it inherited, more diffusely, an expectation among the population that the birr was a stable store of value — an expectation built up over thirty years of imperial monetary discipline that the Derg would inherit, exploit, and ultimately destroy.

The argument of this article is that the destruction was easy because the institutional safeguards were thin. The Derg did not have to break the NBE’s statute to finance its deficits — it only had to use the statute to its full permissive extent. It did not have to abolish the CBE — it only had to extend the CBE’s policy-instrument function to dominate the bank’s commercial function. The institutional door to monetary disorder was already open; the Derg walked through it.

The serious counter-argument

A serious objection to this article’s framing comes from the structuralist tradition in development economics. The objection runs: it is misleading to evaluate the 1945 founding by 2020s standards of central-bank independence, because no such standards existed in 1945. The Bank of England did not become operationally independent until 1997; the Federal Reserve was politically constrained throughout the period; the European Central Bank did not exist. To single out the Ethiopian NBE for being insufficiently independent is to apply an anachronistic standard.⁹

The objection has real force, and the article will not pretend otherwise. The reply is twofold.

First, the article is not claiming that the 1945 founders should have anticipated modern central-bank-independence norms. It is claiming that the institutional features they chose — central-bank discretion to lend to government, dual commercial-and-central-bank function, executive appointment of the governor, statutory limits the government could lift — foreseeably allowed monetary discipline to depend on the restraint of the executive of the day. They did not have to foresee Mengistu’s specific use of those features to know that the features made such use possible.

Second, even by 1945 standards, the Ethiopian arrangement was on the permissive end of the spectrum. The Bank of France of the period had operational independence on monetary issues and a statutory wall between currency issuance and government lending; the Banque Nationale Suisse had stricter limits on lending to the Confederation; Sweden’s Riksbank had been operating with a degree of statutory independence since 1897.¹⁰ The Ethiopian founders were not at the frontier of best practice in 1945. They built a central bank that was institutionally compliant with the executive in a way that even contemporaneous European central banks were not.

This is not an indictment of the 1945 founders. They built what they could build, given the institutional context, and they built something that worked for thirty years. The point is that what they built had a known and foreseeable failure mode, and the failure mode was triggered when an executive willing to use the central-bank-lending mechanism replaced an executive that had chosen not to.

The next article opens Part II of the series with the moment that executive arrived — the 1974 revolution, and the logic of the command economy it imported. The monetary mechanism described here is the mechanism that revolution would exploit.

---

Footnotes

1. National Bank of Ethiopia, “Bank Notes,” https://nbe.gov.et/about-us/bank-notes/: “Not long after the Italian occupation and the attempted transformation of Ethiopia into Italian East Africa, the Italian lira was introduced on 15 July 1936 and Ethiopian banknotes were withdrawn from circulation at 3 lire per talari (birr).” ↩

2. NBE, “Bank Notes”: “In an effort to increase the use of Italian paper money, the exchange rate for silver coins (Maria Theresa thalers) was raised to 4.50 lire, then to 5.00, and eventually, in stages, to 13.50.” ↩

3. Alberto Sbacchi, Ethiopia under Mussolini: Fascism and the Colonial Experience (Zed Books, 1985); Haile M. Larebo, The Building of an Empire: Italian Land Policy and Practice in Ethiopia, 1935–1941 (Clarendon Press, 1994), chapter 5. ↩

4. J. Mwangi S. Kimenyi and Felix Tampuh, “Currency Boards and Currency Systems in Sub-Saharan Africa,” Journal of African Economies 7, no. 1 (1998): 36–62; the East African Currency Board, established in 1919, issued the East African shilling used in Ethiopia between 1942 and 1945. ↩

5. NBE, “About Us,” https://nbe.gov.et/about-us/. ↩

6. NBE, “Bank Notes.” ↩

7. NBE, “About Us.” The 1963 Bank Proclamation No. 207 separated commercial and central-bank functions. ↩

8. Mequanint B. Melesse and Sileshi Yitayih, “The Effects of Currency Devaluation on Ethiopia’s Major Export Commodities,” Cogent Economics & Finance 11, no. 1 (2023). The 2.07 rate was established under Bretton Woods stabilisation in 1949 and held until October 1992. ↩

9. For the standard structuralist case against treating 1990s norms of central-bank independence as universal, see Ha-Joon Chang, Kicking Away the Ladder: Development Strategy in Historical Perspective (Anthem Press, 2002), chapter 3. ↩

10. For the comparative history of central-bank independence, see Christopher Crowe and Ellen E. Meade, “The Evolution of Central Bank Governance Around the World,” Journal of Economic Perspectives 21, no. 4 (2007): 69–90; and Marc Quintyn and Geneviève Verdier, “Mother, Can I Trust the Government? Sustained Financial Deepening — A Political Institutions View,” IMF Working Paper WP/10/210 (2010). ↩