The argument now before the United States Supreme Court is whether a president can dismiss members of the Federal Reserve Board at will. The constitutional question is real, but the more important question is empirical: what does the evidence from monetary history actually say about what happens when central banks come under direct political control?
The answer is clear and it is consistent across a substantial body of evidence.
Turkey offers the most recent controlled experiment. Between 2019 and 2021, President Recep Tayyip Erdogan replaced the central bank governor three times, each removal occurring after the governor declined to lower interest rates in the face of rising inflation. Erdogan held — and publicly expressed — the heterodox view that high interest rates cause rather than reduce inflation. The governors who disagreed were dismissed. The governors who complied lowered rates. Turkish inflation reached 85% in October 2022. The lira lost approximately 80% of its value against the dollar over four years.
The Turkish case is not unique. It is, however, unusually legible. The mechanism is available for inspection: dismiss the governor who disagrees, appoint one who complies, observe the result.
The Turkish case is not unique. It is, however, unusually legible. Argentina’s repeated monetary collapses over three decades consistently trace to episodes in which the central bank’s capacity to resist government pressure was undermined — by direct instruction, by appointment of compliant governors, or by legislation that made it impossible to maintain an independent rate-setting function. In each case, the deterioration of monetary credibility preceded rather than followed the inflationary episode.
The academic evidence on central bank independence and inflation is substantial and directionally consistent. Alberto Alesina and Lawrence Summers published the foundational cross-country study in 1993, finding a strong negative correlation between central bank independence and average inflation rates across OECD countries. Subsequent work by Nergiz Dincer and Barry Eichengreen, updating the dataset in 2014, confirmed the finding with a broader country sample and longer time series. More recent research by Ha, Kose, and Ohnsorge at the World Bank found that central bank credibility — which depends heavily on independence — is a significant determinant of inflation expectations, and that inflation expectations are, in turn, one of the strongest predictors of actual inflation.
The mechanism is not mysterious. Monetary policy requires the ability to make credible commitments over time: a commitment to raise rates if inflation rises, regardless of the short-term political cost. This commitment is credible only if the authority making it cannot be easily overruled by politicians facing electoral cycles that run shorter than business cycles. When markets believe that a central bank’s governor will be removed if they raise rates, they discount the bank’s rate commitments accordingly.
The counter-argument is democratic accountability: monetary policy is too powerful to be insulated from elected oversight. This is a legitimate institutional concern. The evidence from the countries that have managed it best — New Zealand, Canada, Sweden, the United Kingdom after 1997 — suggests that the resolution is not full political control but structured accountability: legislated mandates with clear inflation targets, transparency requirements, and parliamentary rather than executive appointment processes.
The United States built its version of this structure over decades, and the Federal Reserve’s credibility — earned painfully after the Volcker era and the double-digit inflation of the late 1970s — is itself a form of institutional capital. Credibility is not a theoretical asset. It is what allows bond markets to price long-term debt at rates consistent with investment and growth, rather than building in the inflation premium that political monetary regimes historically require.
The argument for central bank independence is not ideological preference for technocracy over democracy. It is structural: the evidence from the countries that have lost it is specific, the mechanism is well-understood, and the costs — measured in inflation rates, exchange rate volatility, and long-term borrowing costs — fall disproportionately on the citizens least equipped to absorb them.
