Base-money growth and inflation in bolívares, 2003–2026: the money printed, and the prices that followed it.
Venezuela · exchange-rate regime

A Currency Is Imported. A State Is Built.

The Kangaroo Peg, thirteen years on: why the regime choice is second order and the deficit and the debt restructuring come first.

Data, with sourceHausmann-model simulationPolicy target, assumptions shown
Introduction

The deficit does not care what the currency is called

Folks confuse the medicine with the symptoms when they ask for dollarization, as if the magic wand of switching to the dollar would cure the deep debt and the fiscal imbalances of the broken state-led model that crippled Venezuela.

But the currency is the second answer, not the first. Ecuador dollarized in 2000. It ended inflation and did not end the deficit, and it defaulted again twenty years later.

The medicine is strong institutions and a privately led economic recovery, which can later lead to dollarization as a way never again to fall into hyperinflation and fiscal imbalance. But dollarization has its own perils and guarantees nothing on its own. In short, we need a legitimate, stable government that can lead a new regulatory framework and create the conditions for a privately led recovery.

In 2013 we argued that a parallel exchange rate with import controls has no steady state unless the public sector's dollar surplus pays for the imports that hold the monetary base still. The choice was reform or hyperinflation. Maduro chose hyperinflation, and it ran from 2017 to 2021: three quarters of output gone, nearly eight million people.

The Kangaroo Peg, Gino Bettocchi and Jesús Bolívar, Second Year Policy Analysis, Harvard Kennedy School, 2013. Advised by Ricardo Hausmann, who coined the term. The paper is not publicly available; the model, data and code behind this update are described at the foot of this page.
−74%output lost, 2013–2020 (real GDP, chained BCV growth)
130,060%peak inflation, 2018 (BCV)
1.1 million %peak exchange premium, Dec 2017 (parallel vs official)
≈8 millionVenezuelans who left the country, 2014–2025
Why not just dollarize? The dollarized twin of the same economy has the identical condition: the deficit must be zero or externally financed. With today's deficit the government's dollar position runs out in about three years; after stages 1–3 it accumulates. The regime choice is second order. The deficit, revenue capacity through private investment, and the institutions that decide how many dollars people want to hold are first order.

The exchange-rate regime was never the disease. It was the symptom of a state under no budget constraint and no obligation to its own law. The chain runs one way: without credible rules there is no private investment; without investment there is no oil and no tax base; without revenue there is a deficit; and a deficit breaks any exchange-rate regime, whether it is denominated in bolívares or in dollars. A currency is imported. A state is built.

That is why 2026 matters. The political transition, the new law on private participation in oil, a transitional government and an election ahead open a window that was shut in 2013: the legal framework can be rebuilt before the deficit is monetised again. The window is short, and the order matters. In 2026 the choice is between private-led growth through institutionalisation and kleptocracy.

Section 1 shows where the country stands in August 2026. Section 2 sets out where it has to be by 2031, with the state balancing its books and restoring the legal framework, private capital doing the investing, and debt exchanged rather than expanded, against assumptions you can move. Section 3 is the sequence. The chart shows the path.

Dollarizing without fiscal reform is a recipe for disaster, 2026–2031

model

The government's dollar position under formal dollarization, in USD billions. Red: dollarize now and change nothing else. The deficit still has to be paid, only now in a currency the state cannot print, and the position is exhausted in about three years. Green: the same dollarization after the fiscal consolidation and the private-capital-led recovery of stages 1–3, which accumulates instead. Same starting reserves, same model, one difference: the deficit.

Hausmann model, dollarized variant (simulate_dollarized in src/model/hausmann.py), August 2026 calibration; starting position is gross reserves. Shown to 2031, the horizon used throughout this piece; the simulation itself runs to 2034, by which point the red path is −$24bn and the green one +$127bn. This is the model's answer to the regime question, not a forecast of whether dollarization happens.
Section 1 · Start here

Where is Venezuela's economy now?

Eight indicators for August 2026, then three charts that carry the paper's argument to today. All values are in constant 2021 bolívares digitales so the three reconversiones (÷1,000 in 2008, ÷100,000 in 2018, ÷1,000,000 in 2021) do not break the lines.

Official vs parallel exchange rate

data

Bs.D per USD, log scale. Solid amber: the main official rate (CADIVI → CENCOEX/DIPRO → DICOM → BCV). Thin dashed: the most depreciated legal window when several coexisted (SITME, SICAD II, SIMADI, DICOM). Blue: the parallel market.

Sources: permuta/lechuga verde quotes 2005–10; DolarToday's own daily history 2010–21 (archived); Monitor Dólar / EnParaleloVzla 2021–26; BCV reference rate; es.wikipedia compilation for decree rates. Shaded bands: reconversiones. Markers: regime changes (hover).

The exchange premium (diferencial cambiario) and the fiscal balance

data

Top: premium = parallel/official − 1, log scale on 1 + premium. Bottom: fiscal balance, % of GDP, annual. The paper's Figure 16 paired these; the premium rose in every year the deficit was monetised.

Premium: assembled series. Fiscal balance: Trading Economics (central government). The 2012 marker is the consolidated public-sector deficit the paper used (15–20% of GDP, incl. PDVSA and FONDEN); no consolidated series is available after 2013, so it is shown as a point, not a line.

Inflation and the exchange premium are driven by money printing, the BCV-funded fiscal deficit

data

In 2012 the imbalance showed up in the exchange premium, a large bolívar base against rationed imports and price controls. Since 2019 it shows up in inflation instead. Both are the same pressure, and the three panels below show where it comes from. Top: base money growth and inflation, both % year on year, log scale. Middle: the bolívar base itself, in constant 2021 bolívares digitales on a log scale, so the three reconversiones do not break the line. Bottom: the same base in US dollars at the parallel rate, the measure that matters for the model: from $15–18 billion in 2010–12 to under $1 billion in 2017–21 and about $2 billion today. A 6% deficit monetised over a $2 billion base produces the inflation of a 20% deficit in 2012.

Right: every month 2004–2026 as one point, base-money growth against inflation (both % YoY, log scales), coloured by year, with the fitted log-log line: . Base money: BCV, base monetaria mensual, all sheets converted to constant 2021 Bs.D (VEB ×10⁻¹⁴, VEF ×10⁻¹¹, VES ×10⁻⁶); growth computed on the converted series. Inflation: BCV via Trading Economics. USD value: base ÷ parallel rate (assembled series).
Section 2 · Where we need to be

From Kangaroo peg to crawling peg to stability and growth by 2031

From state control to private enterprise.

This section shows the way out: public debt down as a share of GDP, the exchange premium closed, and the currency and the economy stabilised. All three turn on the same lever: the state balancing its books and private capital doing the investing. None of them turns on the unit of account. They are targets with their assumptions in view, not forecasts, and every panel recomputes as you move them. And none of them starts on its own: the first step of the roadmap is a legitimate, serious government that restores the legal framework, which is where Section 3 begins.

The development model is private-capital-led. The state does what only the state can do: legitimacy, the legal framework for oil, electricity and property rights, and fiscal balance. Foreign direct investment, repatriated savings and privatisation buyers do the investing. Public borrowing is limited to liability management: old claims exchanged for new bonds. The debt exercise is about reaching a sustainable path, not about financing a state-led recovery.

When people call for dollarization, what they are asking for is economic freedom: an economy led by free Venezuelans. That can be delivered on Benjamin Franklin's paper or on Simón Bolívar's. Which would they really prefer?

Two debt scenarios, both with the same reform programme. A: no haircut, every claim exchanged one for one into new-terms bonds. B: a 50% haircut on the same exchange, credible because it is backed by a legitimate government with the internal factions on board and by the new legal framework.

1. Debt sustainability

targetdata to 2025

Public debt, % of GDP. The bars to 2025 are the Trading Economics series, a broader measure of public liabilities. Scenarios A and B are run on the claims that would actually be exchanged (about $160bn face, the slider), so the dotted grey bridge into 2026 marks a change of definition, not a write-off the policy achieved. The goal is under 100% with interest under 15% of revenue and gross financing needs the state can cover without net new borrowing.

Debt dynamics: Dt = Dt−1 + capitalised interest − overall balance − privatisation proceeds + net new borrowing. Interest is capitalised during the grace period and paid after. The default stock is the face value of exchangeable claims (sovereign ~60, PDVSA ~35, past-due interest ~40, bilateral and arbitration ~25); the Trading Economics ratio of 309% of GDP is a broader definition and is available at the top of the slider.

2. Fiscal balance

targetdata to 2025

Overall and primary balance, % of GDP. Oil revenue up, primary spending down, interest on the restructured debt after grace. Money printing stops when the overall balance is no longer negative.

Primary balance = non-oil revenue (12% of GDP) + oil revenue reaching the state − primary spending (2026: $33bn, cut by the slider over two years, then growing with GDP). Overall = primary − interest paid. Bars to 2025: Trading Economics.

3. Inflation and premium paths

targetmodel

Top: inflation. Bottom: premium. Green dotted: the target glide paths implied by the assumptions (inflation follows money growth less real growth, and any deficit not covered by borrowing forces money growth; the premium follows the fundamentals gap, which oil revenue and FDI dollars close and an unfinanced deficit reopens). Violet dashed: the Hausmann model's status quo and governance-package simulations from the August 2026 calibration.

Model paths from run_model.py: the status quo keeps the deficit monetised and the BCV rationing FX; the governance package cuts spending, lifts oil revenue on the same path as the slider default, and lowers the desired dollar share. The model has no price-level block, so inflation is target-only. Model premium is shown on the same log scale.

4. Five-year paths: premium and the nominal exchange rate

targetmodel

Top: the exchange premium on a plain (not log) scale, year by year, under each path. Bottom: where the official rate could go in nominal terms from 806 Bs/USD in September 2026, year by year to 2031, under the same paths. The rate axis is logarithmic only because the status-quo path runs into the millions; the premium axis is linear as requested.

5. Exchange-rate regime timeline

target

Managed rate with controls → unification → float with bands. The float year is computed: the first year the premium is under 10% with no deficit to monetise, inside the end-2027 to latest-year window; it turns red if it has to be forced.

AssumptionValueBasis
Section 3 · The sequence

The policy path, explained

Strip the 2013 paper down and one question is left: after the state has paid for its own imports, does it have enough oil dollars to cover what the rest of the country needs to import? Whatever is missing, the central bank prints. That is the whole mechanism. In 2005 the oil dollars covered it and the system held, with a premium of 26%. By 2012 the state needed nearly four times what it had, and no devaluation could close that: the premium went to 305%, and from there to a million. In August 2026 the shortfall is about $5 billion a year, roughly the size of the fiscal deficit.

The five stages below each close part of that shortfall, and they only work in that order. Stage 1 is the government itself: without a legal framework nobody brings capital, so stage 2 has nothing to ramp. Stage 3 stops the printing, and until it does the premium cannot stay down in stage 4. Stage 5, floating the bolívar, is less a decision than an announcement: you declare it once the market has already done it.

Scroll sideways to see the whole path.

STAGE 1 · 2026–27 Legitimacy &legal framework Oil, electricity, property lawsDebt exchange opened → unlocks FDI, new-terms debt STAGE 2 · 2026–30 Private-capitalproduction ramp FDI, JV oil output, concessionsBuyers bring capital, not the state → the dollar gap closes STAGE 3 · 2027–28 Fiscal surplus,printing stops Spending cut, arrears exchangedBase money grows with GDP → fd ≤ 0, debt/GDP falling STAGE 4 · 2027–29 Disinflation,premium closes Inflation follows moneyPremium follows the gap → premium < 10% for 12 months STAGE 5 · 2027–29 Float, controlsremoved Managed float with bandsDe jure follows de facto → premium at zero TIMELINE

The premium and the rate, month by month, 2003 to 2031

targetmodeldata to Aug 2026

The payoff of the sequence, month by month. Top: exchange premium. Bottom: bolívares per US dollar, official and parallel. Data to August 2026, then a projection guided by Section 2. The premium follows chart 3's target path: about 19% today, collapsing toward zero as the fundamentals gap closes and unified by the float year, with the Hausmann governance-package path in dashed violet. The official rate follows chart 3's inflation path under purchasing-power parity, less the real appreciation typical of a stabilisation, calibrated so the 2027 average is about 1,200 Bs/USD and moving around that level thereafter.

Data: assembled official and parallel series. Projection: monthly inflation interpolated (log-linear) between today's pace and chart 3's yearly inflation path; official rate = previous month × (1 + monthly inflation − US inflation) × (1 − real appreciation) until end-2027, then pure PPP drift; parallel = official × (1 + premium). Model path from run_model.py, aligned at August 2026.