Nationalisation and the Printing Press: The Derg's Monetary Regime

What “nationalisation of the banks” actually did

The January 1975 nationalisation of Ethiopia’s banks is usually narrated as a social-policy event — a transfer of ownership from foreign and domestic capital to the people of Ethiopia. This article argues that its consequences were primarily monetary, and that the monetary consequences set up everything that followed: the inflation, the famine, the parallel exchange rate, the chronic forex shortage that would dog Ethiopia for the next forty years. The bank nationalisations were not merely a redistribution of ownership. They were a structural change in how money got created and where credit went, and the structural change is the part this article is about.

The thesis is this: when the state owns the banking system and the same state runs a deficit it has chosen not to finance through taxation, the banking system’s deposits become the available funding source. The mechanism by which they become the funding source is central-bank credit to the government, intermediated through the state-owned banks. The technical name for this mechanism is fiscal dominance — a situation in which monetary policy is, in effect, determined by the fiscal position rather than by inflation targeting. The Derg’s banking architecture made fiscal dominance not an emergency arrangement but the default operating mode of the monetary system. Everything that follows in this series — in some form, under every subsequent regime — is downstream of that choice.

The architecture

Proclamation 26 of January 1975 nationalised the three foreign-owned banks operating in Ethiopia (Addis Ababa Bank, Banco di Roma, Banco di Napoli) and merged them with the State Bank of Ethiopia and the Commercial Bank of Ethiopia into a consolidated state banking system.¹ The same proclamation nationalised the insurance companies, folding them into a single Ethiopian Insurance Corporation. By the end of 1975, the financial sector consisted of three institutions: the National Bank of Ethiopia (central bank), the Commercial Bank of Ethiopia (commercial banking), and the Development Bank of Ethiopia (which had absorbed the former Agricultural and Industrial Development Bank).² All three were owned by the state. All three reported, ultimately, to the same political authority.

Two further proclamations completed the architecture. The Agricultural and Industrial Development Bank Proclamation of 1979 expanded the DBE’s mandate to provide directed credit to state farms, cooperatives, and state-owned industrial enterprises at administered interest rates lower than those available to private borrowers (where private borrowers existed at all).³ And the establishment of the Housing and Savings Bank in 1975, later renamed the Construction and Business Bank, gave the state another vehicle for mobilising household savings into state-directed lending.⁴

The effect of all this was that household deposits — the savings of urban Ethiopians, accumulated in state banks because there was nowhere else to put them — flowed, through the state banking system, into lending decisions made by political authorities. The depositors had no say in how their money was lent. The borrowers were, overwhelmingly, state-owned enterprises and state-financed projects. Private borrowing existed but at the margins: small traders, a residual private sector grudgingly tolerated, with credit allocations that were small, expensive, and politically conditioned.

This is what financial repression looks like in its mature form. The depositor earns an administered interest rate set below the rate of inflation, so the real value of his deposit declines while he holds it. The state borrower receives credit at an interest rate set below what a market would charge, with no expectation of being held to commercial discipline by the lender (because the lender is, ultimately, the same state). The difference between the rate the depositor earns and the rate the borrower pays — the spread, in normal banking — becomes a tax on savings, paid into the state’s accounts, that funds state priorities without ever appearing in a budget anyone gets to vote on.

The deficit and the printing press

The fiscal position the Derg accumulated through the late 1970s and 1980s was, by any honest accounting, unsustainable. The regime fought four simultaneous wars (against the Eritrean independence movements, against the TPLF and other Tigrayan and Oromo insurgents, against Somalia in the Ogaden, and intermittently against various smaller resistances), maintained one of the largest standing armies in sub-Saharan Africa, ran a sprawling state-enterprise sector that consistently lost money, and provided urban food subsidies through the AMC at prices that did not cover procurement and distribution costs.

By the mid-1980s, Ethiopian official sources acknowledged that military expenditure absorbed approximately 46 percent of central government expenditure; some independent estimates were higher.⁵ Civilian expenditure on health fell from around 6 percent of the budget in 1973–74 to around 3 percent by 1990–91.⁶ The state-enterprise sector required continuous subsidies; the AMC’s grain-procurement-and-distribution operation required continuous subsidies. Total tax revenue, by contrast, was limited by the underlying tax base: the Derg’s land reform had eliminated the rural land tax that the empire had relied on, the nationalisations had eliminated the private-sector profits that could be taxed, and the urban income-tax base was too small to fund the regime’s ambitions.

The mathematics of this situation force a specific outcome. If revenue cannot rise to meet expenditure, expenditure must fall to meet revenue, or the gap must be filled by borrowing. The Derg was not willing to cut its military, its state enterprises, or its food subsidies; foreign borrowing was constrained by Western sanctions and by Soviet aid that came mostly as military equipment rather than convertible currency; domestic borrowing through bond markets was not an option in a country that had nationalised its banks and had no bond market to speak of. The only remaining mechanism was borrowing from the central bank — that is, central-bank credit to the government, which is the formal mechanism by which monetary expansion finances fiscal deficits.

The NBE’s lending to the central government, modest through the early 1970s, expanded sharply through the late 1970s and 1980s. By the late 1980s, NBE advances to the government had become the dominant source of base-money expansion in the economy.⁷ The money supply (M1) grew at rates that, in some years, exceeded the rate of nominal GDP growth by a significant margin — the textbook signature of monetary financing of deficits.

The lag, and the disguise

Here is the part of the story worth understanding because it explains why the Derg got away with this longer than seems possible. Inflation, when it is created by monetary expansion, arrives with a lag. The textbook estimate is six to eighteen months, depending on the structure of the economy and the speed of price adjustment.⁸ In an economy with extensive price controls, the lag is longer still, because the inflation cannot show up in the prices the state controls. It shows up first in the prices the state does not control: the parallel exchange rate, the open-market price of food in regions where the AMC’s grip is loose, the prices of imported manufactured goods that have circumvented the official import system.

This is the mechanism by which the Derg’s monetary policy looked successful in the short term. The official inflation rate, calculated against a basket of mostly controlled prices, stayed low through much of the 1970s and into the early 1980s. The official exchange rate remained 2.07 birr/USD throughout the period. The state’s official statistics told a story of monetary stability that was, in narrow accounting terms, true.

What was true at the same time was that the unobserved prices were moving. The parallel exchange rate, where it could be measured, was rising sharply through the late 1970s and 1980s — by some estimates it reached 6 to 8 birr per dollar by the mid-1980s, against the official 2.07.⁹ The open-market price of teff in Addis Ababa was, by 1984, more than four times the AMC’s official farmgate price.¹⁰ The disparity is documented in the Wikipedia summary of Ethiopian agricultural pricing: the 1984–85 official procurement price for 100 kilograms of teff was 42 birr at the farm level and 60 birr when the AMC purchased it from wholesalers, while the same quantity retailed at 81 birr at urban kebele food stores and sold for as much as 181 birr in the open market.¹¹ The price system was, in other words, transmitting the information about underlying scarcity through the channels the state could not suppress. The state was not seeing it, or was choosing not to see it.

The honest counter-argument

A serious case must engage the strongest version of the Derg’s defence, which is this: the regime was facing exceptional fiscal pressures — wars on multiple fronts, sanctions, a hostile international environment, droughts — and any government in its position would have used the central bank to finance its operations. To single out the Derg for monetary mismanagement is to apply a peacetime standard to a wartime government.

The argument has real force, and the series does not dismiss it. Three replies, in increasing seriousness.

First, the wars were not entirely external impositions on the regime. The Eritrean conflict had escalated after the Derg’s decision to abandon federation and pursue a military solution; the Tigrayan and Oromo insurgencies were responses to the Derg’s own policies; the Ogaden war was not something the Derg sought, but the post-war military build-up the Derg maintained was a choice. To treat the wars as exogenous shocks to which the regime had to respond is to leave out the regime’s role in producing them.

Second, other regimes facing comparable fiscal pressures have not necessarily resorted to monetary financing. Israel through the 1970s and 1980s, fighting multiple wars and absorbing large immigration shocks, maintained more disciplined monetary policy through institutional commitments. Vietnam after 1986, recovering from a much longer war than the Derg fought, deliberately avoided the central-bank-financing route after its 1980s hyperinflation experience. The choice to finance deficits monetarily is a choice, even under fiscal pressure. The Derg made it consistently.

Third — and this is the deepest reply — the Derg’s monetary policy was not a temporary wartime expedient. It was the default operating mode of the regime across its entire seventeen-year existence, including in years when the war situation was relatively stable. The inflation that arrived in the 1980s was not the cost of a brief emergency; it was the cumulative cost of a sustained policy. Treating the Derg’s monetary regime as wartime expediency is too generous a framing for a regime that ran monetary policy this way in peace as much as in war.

What the Derg left behind

The collapse of the Derg in 1991 left the EPRDF, which inherited the regime’s banking system, central bank, monetary statutes, state enterprises, and the 2.07 official exchange rate that had ceased to bear any relationship to economic reality. Inflation in the final years of the Derg ran at rates the official statistics did not capture accurately; the parallel exchange rate was probably between 6 and 10 birr per dollar; the country had accumulated significant external debt that it was effectively unable to service.¹²

The choice the EPRDF faced was binary in principle: either reform the monetary architecture — change the central-bank statute, recapitalise the banks, devalue the official rate to something near the parallel rate, commit to fiscal discipline — or keep the architecture and run it differently. The EPRDF announced it would do the first. It mostly did the second. The 1992 devaluation, examined in article 9, acknowledged reality just enough to get IMF and World Bank support; it did not change the underlying mechanism. The central bank remained statutorily able to finance the government; the state banks remained dominant; financial repression remained the operating mode. The wallpaper changed. The walls stayed up.

The next article turns to what happened when the suppressed prices and the unfinanced subsidies and the printed money all met a drought, in 1984.

---

Footnotes

1. Proclamation No. 26/1975, “A Proclamation to Provide for the Government Ownership of Banks, Insurance Companies and Financial Houses,” 1 January 1975. Discussed in Eshetu Chole, Underdevelopment in Ethiopia (OSSREA, 2004), chapter 5. ↩

2. National Bank of Ethiopia, “About Us — History,” https://nbe.gov.et/about-us/; Befekadu Degefe and Berhanu Nega, eds., Annual Report on the Ethiopian Economy, Vol. 1 1999/2000 (Ethiopian Economic Association, 2000). ↩

3. Proclamation No. 158/1979, on the Agricultural and Industrial Development Bank. Befekadu and Berhanu, Annual Report on the Ethiopian Economy, chapter 4. ↩

4. NBE, “About Us — History.” ↩

5. “1983–1985 famine in Ethiopia,” Wikipedia, https://en.wikipedia.org/wiki/1983%E2%80%931985_famine_in_Ethiopia, citing Mengistu Haile Mariam’s own 1984 announcement that 46 percent of GNP would be allocated to military spending — a claim that conflates GNP and government expenditure but is widely repeated. ↩

6. “1983–1985 famine in Ethiopia,” citing the same figures for the health budget. ↩

7. National Bank of Ethiopia, Quarterly Bulletins, 1980s issues, summarised in Befekadu and Berhanu, Annual Report on the Ethiopian Economy, chapter 6. ↩

8. For the standard estimates of the monetary-policy-to-inflation lag, see Ben S. Bernanke and Frederic S. Mishkin, “Inflation Targeting: A New Framework for Monetary Policy?”, Journal of Economic Perspectives 11, no. 2 (1997): 97–116; and Lars E.O. Svensson, “Inflation Targeting,” in Handbook of Monetary Economics, Vol. 3B (2011). ↩

9. For estimates of the Derg-era parallel exchange rate, see Alemayehu Geda, “The Macroeconomic Performance of the Ethiopian Economy,” in Befekadu and Berhanu, Annual Report on the Ethiopian Economy, Vol. 1. ↩

10. “Agriculture in Ethiopia,” Wikipedia, https://en.wikipedia.org/wiki/Agriculture_in_Ethiopia, citing the 1984–85 official procurement prices versus open-market prices. ↩

11. “Agriculture in Ethiopia”: “the 1984–85 official procurement price for 100 kilograms of teff was 42 birr at the farm level and 60 birr when the AMC purchased it from wholesalers. But the same quantity of teff retailed at 81 birr at food stores belonging to the urban dwellers’ associations (kebeles) in Addis Ababa and sold for as much as 181 birr in the open market.” ↩

12. For end-of-Derg debt and inflation figures, see World Bank, Ethiopia: Toward Poverty Alleviation and a Social Action Program (World Bank, 1993). ↩