Zimbabwe Printed Money While Using the US Dollar. Inflation Followed.

Every currency regime rewards someone. Zimbabwe's 2009 dollarization took a reward away from the treasury: the option to pay for a deficit with money the Reserve Bank of Zimbabwe (RBZ) created. That option is the real thread through Zimbabwe's monetary history since 2008. Hyperinflation ended when the treasury lost it. Inflation came back when the treasury found it again, and the economy was still legally dollarized when that happened. From 2014 to 2018 the government created "dollars" the RBZ could not redeem, as bank balances and bond notes held at a fictional one-to-one parity, and the fiction broke in 2019. The currency label mattered less than the budget constraint. The 2024 gold-backed currency (the ZiG) repeated the pattern within six months under a single label.

So my thesis is narrower than the working version I started with. Dollarization did not fail in 2016 to 2019. It was hollowed out from inside. A country can use the dollar and still print, provided its central bank can create claims that are called dollars. Zimbabwe did exactly that. Ecuador came uncomfortably close.

How you end a hyperinflation without a new central banker

The scale first. Zimbabwe's official statisticians reported annual inflation of 11.2 million percent for June 2008, and Steve Hanke's estimate for mid-November 2008 was 89.7 sextillion percent year on year, or 790.6 billion percent month on month [2]. Those are guesses at a number nobody could measure any more. The government had stopped publishing inflation figures in July 2007 [2]. Put that on my list of official statistics that were quietly redefined. This one was redefined as silence.

The mechanism is not in dispute. The IMF's 2019 Article IV report (the Fund's annual review of a member economy) describes it in one line: high fiscal deficits financed by RBZ money printing led to hyperinflation and the end of the original Zimbabwe dollar [1]. The RBZ also ran large "quasi-fiscal" programs, meaning budget spending done by the central bank so that it never shows up in the budget. The IMF's 2009 consultation blamed those activities for the hyperinflation and for a fall in real output estimated at 14 percent in 2008 [3].

Then the local currency simply disappeared. By October and November 2008 it was barely circulating. In February 2009 the authorities made the multicurrency system official [3], and printing of the Zimbabwe dollar stopped in April 2009 [2]. The government did two other things at the same time, and the order matters. It committed to eliminate the RBZ's quasi-fiscal activities and to run a cash budget, matching each month's spending to that month's revenue [3]. Finance Minister Tendai Biti's slogan was "we eat what we kill." An IMF departmental paper from 2010 credits dollarization with stabilizing prices and imposing fiscal discipline, "including on the RBZ" [4].

The World Bank's consumer price series for Zimbabwe has no value for 2005 to 2009, which is honest, since nobody could say what a meaningful annual rate was. It restarts at 3.02% in 2010, then 3.47% in 2011, 3.73% in 2012 and 1.63% in 2013, followed by three years of mild deflation: -0.20% in 2014, -2.43% in 2015 and -1.54% in 2016 [5]. The budget numbers fit the same story. Deficits ran between 0.35% and 1.98% of GDP every year from 2009 to 2014 [6].

The consensus reads this as a victory for the dollar. I offer it my mock-solemn congratulations. The dollar did end the hyperinflation. But the dollar worked through one channel: it took away the RBZ's printing press at the same moment the treasury agreed to a cash budget. Both happened in 2009, so the 2009 episode alone cannot separate them. The next ten years can.

How Zimbabwe printed money without a currency

Look at what happened while the dollar was still legal tender. The budget deficit rose from 1.61% of GDP in 2015 to 4.99% in 2016 and 8.68% in 2017, then fell to 4.55% in 2018 [6]. (Other published series put 2017 higher. The denominator is a problem in itself: nominal GDP in 2018 and 2019 depends on which exchange rate you use to convert the RTGS figures, and that rate was the very thing in dispute.) The spending went mainly to farm subsidies and above-market grain purchases, at a time when revenue fell with commodity prices [7].

A dollarized government that cannot borrow abroad should have to cut spending. Zimbabwe's government found another way. It sold Treasury bills (short-term government debt) to domestic banks, and when banks would not take any more, the RBZ gave it an overdraft. By the IMF's account in July 2017, financing needs had outrun the private sector's appetite for T-bills, so the government drew on the RBZ, and domestic debt reached almost 25% of GDP at end-2016, up from 14% in 2015 [7]. Executive directors warned that central bank credit to the government "had significant potential for generating inflationary pressures" [7].

Here is where the statistic gets interesting. The money created this way was not paper Zimbabwe dollars. It sat in bank accounts as balances in the real-time gross settlement system (RTGS, the interbank electronic payment system), and those balances were denominated in US dollars. They were not backed by US dollars. T-bills outstanding rose from US$2.1 billion in 2016 to a cumulative US$7.6 billion by August 2018, from 4.4% of GDP in 2014 to 36.5% [9]. RTGS money grew from under 1 billion in 2016 to more than 9 billion in December 2018 [10]. The physical side was bond notes, launched on 28 November 2016, supposedly backed by a US$200 million facility from the African Export-Import Bank and declared equal to the US dollar [8].

That one-to-one parity is a legal fiction, and it shaped the CPI. A price index measures prices in the unit people quote. If shops quote in "dollars" and the official rule says an RTGS dollar is a dollar, the index records stable prices while the money people actually hold loses value. Zimbabwe's measured inflation was 0.89% in 2017 and 10.62% in 2018 [5]. Meanwhile bond notes traded at about 3 to the US dollar by January 2019 [8]. That is a loss of $1 - 1/3 \approx 67\%$ of the notes' value against the dollar (my arithmetic, not Lab output). The inflation had happened. It showed up as a parallel-market premium, as cash queues, and as shops posting different prices for cash dollars, bond notes and card payments, not as CPI. On 1 October 2018 the authorities ordered banks to split foreign currency accounts into "Nostro" accounts backed by real dollars and "RTGS" accounts, while still insisting on one-to-one [9]. That was a regulator admitting there were two currencies and asking everyone to pretend there was one.

The IMF summary of the whole episode is blunter than I would dare to be. The resumption of large deficits financed by non-convertible quasi-currency instruments created distortions that forced the authorities to abandon the dollarized system and adopt a new domestic currency in early 2019 [1].

February 2019: the bill arrives

The RTGS dollar became official on 25 February 2019 at 2.5 to the US dollar. On 24 June 2019 it was renamed the Zimbabwe dollar and foreign currencies were banned [11]. Annual inflation passed 100% in June 2019, passed 500% in December 2019 and peaked at 837.53% in July 2020 [11]. On the World Bank's annual series that is 255.30% for 2019 and 557.20% for 2020 [5]. The currency ended at 30,000 per US dollar on 5 April 2024 [11]. From 2.5 to 30,000 is a factor of 12,000 in five years (again my arithmetic).

The obvious reading is that inflation came back when Zimbabwe left the dollar, so the dollar was what had been holding prices down. I think that reading has the timing backwards. The money was created in 2014 to 2018, under the dollar. In February 2019 the authorities stopped pretending about its value.

The strongest objection: 2019 had a balanced budget

This is the fact that should worry me most, so here it is at full strength. Finance Minister Mthuli Ncube's Transitional Stabilisation Programme (October 2018 to December 2020) included a 2-cents-per-dollar tax on electronic transfers [9]. The budget went from a 4.55% deficit in 2018 to roughly balance in 2019 (-0.06% of GDP) and a small surplus in 2020 (+0.53%) [6]. If fiscal dominance is the story, inflation should have stopped in 2019. Instead it ran to 521% by year-end [1]. A critic can say the budget was closed and prices exploded anyway, so it was the currency change, the float and the loss of the dollar anchor that did the damage. That critic has a point. The float did the damage in the sense that it released it.

My answer has two parts, and both come from the IMF's own account of 2019.

First, a stock, not a flow. More than 9 billion of RTGS balances had been created at a one-to-one rate that nobody believed [10]. Once that money could be priced honestly, its value against the dollar had to fall to whatever the dollar backing could support, and prices quoted in it had to rise. The IMF attributes the end-2019 inflation to exchange-rate depreciation driven by the "monetary overhang of past monetary financing of deficits" [1]. Closing the deficit in 2019 stopped new money from being created. It could not shrink the money already out there.

Second, the printing never fully stopped. It moved off the budget. Reserve money nearly tripled in the second half of 2019, from ZWL$3.3 billion at end-June to an estimated ZWL$9 billion at year-end, which the IMF attributes to the resumption of the RBZ's quasi-fiscal operations [1]. The budget line said balance. The central bank's balance sheet said otherwise. This is my blind spot to watch: I can make a story fit almost any outcome by moving spending between the budget and the central bank. So here is the test that could have proved me wrong. If 2019 inflation had come with flat reserve money and no quasi-fiscal activity, the currency-anchor view would win. It did not come that way.

I will also concede part of the objection. The regime did matter, because the change in regime is what turned a hidden loss into an open one. A pure fiscal story does not explain why the price level jumped in 2019 rather than 2017. The currency regime decided when Zimbabweans paid. The financing decided whether they paid.

2024: one currency, two halves of a year

The cleanest natural experiment came last, and nobody planned it. The ZiG (Zimbabwe Gold) launched on 8 April 2024 at 13.56 per US dollar, backed by about US$1.6 billion in gold and foreign reserves [12]. Same backing story, same currency name, same central bank. From April to September 2024 the monetary base grew 215%, "driven by RBZ financing," according to the IMF's 2025 Article IV consultation as summarized by EquityAxis [13]. On 27 September 2024 the RBZ devalued the ZiG by 42.55%, and that month prices rose 37.2% in ZiG terms against 0.7% in US dollar terms [12]. Monthly ZiG inflation reached 20% in October 2024 [13].

Then the financing stopped. The IMF welcomed "the halting of quasi-fiscal operations and monetary financing" and the transfer of those operations to the Treasury [13]. Monthly ZiG inflation fell to 0.5% by February 2025 [12] and to 0.3% by June 2025, and the gap between official and parallel exchange rates narrowed from 50% to 20% by September 2025 [13]. The currency and the gold were the same from April 2024 to June 2025. The financing was what changed.

Now the part the victory lap leaves out. The deficit did not vanish. The same IMF review puts the fiscal deficit at 4.5% of GDP in both 2023 and 2024, with domestic expenditure arrears rising from about $200 million at end-2023 to nearly $600 million at end-2024 [13]. When the printing stopped, the government stopped paying some of its bills instead. That is still a budget constraint at work, but someone else carries the cost.

The Ecuador counterexample I owe myself

In my post on Ecuador and El Salvador I credited dollarization with ending Ecuador's inflation. Zimbabwe makes me add a condition. Dollarization ends inflation only if the central bank cannot create dollar-denominated claims on itself to pay the treasury's bills. Ecuador shows the condition is not automatic. A 2014 monetary code allowed Ecuador's central bank (the BCE) to finance the government directly. After the 2014 oil price collapse it did: credit to the government reached about 7% of GDP in 2017, and the BCE's reserve coverage of bank deposits fell below 40%, from above 100% in the early 2010s [14]. Those deposits were dollars on paper. Below 40% coverage, they were heading toward RTGS dollars. Ecuador stopped before it got there. In April 2021 its National Assembly passed a law barring the BCE from "directly or indirectly" financing the state [15], a condition of a $6.5 billion IMF program.

So the dollar is not the commitment device. The commitment device is a rule that the central bank cannot lend to the government, and a government willing to obey it. Ecuador needed a statute to rebuild that rule twenty years after it dollarized. Zimbabwe kept the dollar, lost the rule, and got inflation anyway. I am raising my confidence that fiscal dominance (a government's financing needs dictating monetary policy) matters more than formal currency arrangements, from 0.7 to 0.75. Two countries are not a sample. But the Zimbabwe ordering, with money created first and inflation recorded later, is hard to get from a currency-anchor model.

A forecast, and who pays

If I am right, the ZiG's future depends on whether the RBZ stays out of government finance, not on the gold in its vault. My sources stop at September 2025, so here is a conditional forecast I can be scored on. I put 0.3 on the RBZ's official ZiG interbank rate being weaker than 40 ZiG per US dollar on 2027-09-30, as published by the Reserve Bank of Zimbabwe. That would be a depreciation of more than a third from the 25 to 27 range it held from November 2024 [12]. A move that size would most likely mean monetary financing had resumed. If it happens with flat reserve money, I am wrong about the mechanism.

The policy that follows is not "dollarize" and not "back the currency with gold." It is a hard, audited ban on central bank credit to the government, direct or through Treasury bills the central bank discounts, with quasi-fiscal programs moved onto the budget where parliament can see them. Ecuador wrote that ban into law in 2021. Zimbabwe moved the RBZ's quasi-fiscal operations to the Treasury in 2024 to 2025, but the arrears show the deficit is still looking for a way out.

Who pays for it? Under the ban, the people who were paid with new money: farmers selling grain to the state above market prices, importers getting foreign currency at below-market rates, and suppliers to the government who are now waiting in a $600 million arrears line [13]. Without the ban, the cost falls on everyone holding the local money, and the people least able to hold dollars pay the most.