Category Archives: Contributions

“The Political Economy of Early COVID-19 Interventions in US States,” CEPR, 2022

CEPR Discussion Paper 16906, January 2022, with Martin Gonzalez-Eiras. PDF (local copy).

We investigate how politico-economic factors shaped government responses to the spread of COVID-19. Our simple framework uses epidemiological, economic and politico-economic arguments. Confronting the theory with US state level data we find strong evidence for partisanship even when we control for fundamentals including the electorate’s political views. Moreover, we detect an important role for the proximity of elections which we interpret as indicative of career concerns. Finally, we find suggestive evidence for complementarities between voluntary activity reductions and government imposed restrictions.

Forthcoming in the JEDC.

Interview, Riksbank RN, 2021

Riksbank Research News 2021, December 2021. PDF (pp. 2–3), HTML.

Q: You have been leader of the CEPR Research and Policy Network on FinTech and Digital Currencies since 2021 and explored issues at the heart of monetary theory and payment systems in your research. What do you think is new about digital central bank money and what makes it different from other digital means of payment?

A: Societies have been using digital means of payment for decades. Commercial banks use digital claims against the central bank, “reserves,” to pay each other. Households and firms use digital claims against commercial banks, “deposits,” as well as claims on such deposits, as money. Financial innovations typically improved the convenience for users or helped build additional layers of claims on top of each other, fostering fractional reserve banking and raising money multipliers.

Recently, new digital instruments have appeared on the fringes of the financial system. Some think of them as currencies and others as mere database entries. These instruments exploit the fact that smart ways of managing information, and even smarter approaches to providing incentives in anonymous, decentralized networks can replicate some functions of conventional monies. Monetary theorists are not surprised. They have debated for decades to what extent money is, or is not a substitute for a large societal database. The information technology revolution has made this debate much less theoretical.

Of course, the new entrants such as Bitcoin have not been very successful so far when it comes to actually creating substitute monies. But they have been quite successful in terms of creating new assets, mostly bubbles. Bubbles are also a great mechanism for their creators to extract resources from other people.

What is new about digital central bank money for the general public (central bank digital currency, CBDC) is that households and firms would no longer be restricted to cash when they wanted to pay using a central bank (i.e., government) liability. That is, banks would lose a privilege and households and firms would gain an option. CBDC, which I like to think of as “Reserves for All,” seems natural when you consider the history of central banking. It also seems natural when you consider that many governments strongly discourage the use of cash. Nevertheless, compared with the status quo, “Reserves for All” would amount to a major structural change.

Q: What do you think are the main challenges of issuing a CBDC?

A: From a macroeconomic perspective, introducing “Reserves for All” could have major implications. The balance sheets of central banks would likely expand while commercial banks would likely lose some deposits as a source of funding. Mechanically, they would reduce their asset holdings or attract other sources of funding. The question is, which assets they would shed, and subject to which terms and conditions they would attract new funding. These are important questions because banks play a key role in the transmission of monetary policy to main street.

While many central bankers are concerned about the implications of CBDC for bank assets and funding costs academic research conveys a mixed picture. To assess the consequences of “Reserves for All” it is natural to first ask what it would take to perfectly insulate banks and the real economy from the effects of CBDC issuance. As it turns out, the answer is “not much:” Under fairly general conditions the central bank holds a lot of power and can neutralize the implications of CBDC for macroeconomic outcomes.

Of course, central banks might choose to implement other than the neutral policies. In my view, this is in fact very likely, for reasons related to the political economy of banking and central banking. On the one hand, CBDC would make it even harder for central banks to defend their independence. On the other hand, CBDC would increase the transparency of the monetary system and trigger questions about the fair distribution of seignorage. On top of this, “Reserves for All” might trigger demands for the removal of other “bank privileges:” Interest groups might request LOLR-support, arguing that they are systemically important and just temporarily short of liquidity. Others might want to engage in open market operations with the central bank.

Beyond macroeconomics and political economy, CBDC could substantially change the microeconomics of banking and finance. In the current, two-tiered system there is ample room for complementarities between financing, lending, and payments. The information technology revolution strengthens these complementarities but it also generates new risks or inefficiencies. How the connections between money and information currently change is the subject of ongoing research. I don’t think we have been able to draw robust conclusions yet as to what role CBDC would play in this respect.

Q: Should we, and will we have CBDCs in the near future?

A: Some countries have already decided in favor. Others, like the Riksbank I believe, are still on the sidelines, thinking about the issues, watching, and preparing. Yet others have only recently taken the issue more seriously, mostly because of the Libra/Diem shock in June 2019, which made it clear to everybody that the status quo ceases to be an option.

I think the normative question is still unanswered. Not only does CBDC have many consequences, which we would like to better understand. There are also the unknown consequences that we might want to prepare ourselves for. Moreover, many of the problems that CBDC could potentially address might also allow for different solutions; the fact that CBDC could work does not mean that CBDC is the best option.

In a recent CEPR eBook* several authors share that view, which suggests a case-by-case approach. CBDC might be appropriate for one country but not for another, for instance because cash use has strongly declined in Sweden and this may favor CBDC (as Martin Flodén and Björn Segendorf discuss in their chapter) while the same does not apply in the US or elsewhere.

Regarding the positive question, I think that many more countries will decide to introduce “Reserves for All,” and quite a few of them in the next five years. One reason is that it is politically difficult to wait when others are moving ahead. Another is the fear of “dollarization,” not only in countries with less developed financial markets. The strongest factor, I believe, is the fear that central banks might lose their standing in financial markets. This is connected with the important question, which the Riksbank has been asking early on, whether in the absence of CBDC declining cash circulation could undermine trust in central bank money.

Among the eBook authors, most but far from all expect that a CBDC in a developed economy would resemble deposits in terms of user experience. Almost everyone expects that private banks and service providers rather than the central bank itself would interact with end-users. I share these views. But there is disagreement as to whether digital currencies would be interest bearing and how strictly they would protect privacy. I believe that it is also unclear how strictly central banks would enforce KYC regulation or holding restrictions on foreigners. These two factors might critically affect the threats to monetary sovereignty in other countries, and as a consequence they might shape the chain reaction of adoptions.

What seems clear to me is that the implications of CBDC go far beyond the remit of central banks. Parliaments and voters therefore should have the final say.

* Dirk Niepelt (2021), editor: “CBDC: Considerations, Projects, Outlook”, CEPR eBook. Changes in the research staff

“Digitales Notenbankgeld – und nun? (CBDC—What Next?),” FuW, 2021

Finanz und Wirtschaft, December 8, 2021. PDF.

  • I draw some conclusions from the CEPR eBook on CBDC, namely:
  • Banks will change, whatever happens to CBDC.
  • The main risk of retail CBDC is not bank disintermediation.
  • CBDC may not be the best option even if it has net benefits.
  • It should be for parliaments and voters, not central banks, to decide about the introduction of CBDC.

“Reserves for All: Political Rather Than Macroeconomic Risks,” CEPR, 2021

Chapter 5 in the CEPR eBook, November 24, 2021. HTML.

From the conclusion:

From a macroeconomic perspective, central banks can largely neutralise the consequences of CBDC. What is highly uncertain, however, is whether they would choose to do so – the political risks of ‘Reserves for All’ are first-order. The decision for or against CBDC thus should not only reflect the assessment of economic trade-offs, but also whether societies are confident in their ability to efficiently manage conflicts of interest. If not, and if they fear that the introduction of CBDC could further politicise banking and central banking, then the introduction of CBDC might constitute a risky regime change. It will be interesting to see how different [countries] judge this risk.

“CBDC: Considerations, Projects, Outlook,” CEPR/VoxEU, 2021

CEPR eBook, November 24, 2021. HTML.

VoxEU, November 24, 2021. HTML.

Retail central bank digital currency has morphed from an obscure fascination of technophiles and monetary theorists into a major preoccupation of central bankers. Pilot projects abound and research on the topic has exploded as private sector initiatives such as Libra/Diem have focused policymakers’ minds and taken the status quo option off the table. In this eBook, academics and policymakers review what we know about the economic, legal, and political implications of CBDC, discuss current projects, and look ahead.

“Die Nationalbank ist an vielen Fronten gefordert (Challenges for the Swiss National Bank),” NZZ, 2021

NZZ, August 10, 2021. PDF (title changed by NZZ). Related article in Ökonomenstimme. HTML.

Should the SNB follow the Fed and the ECB and rework its strategy? There is a case for rethinking the broad inflation target, the monetary policy concept, and the communication strategy. Equally important is a strategy review outside of the SNB: The SNB cannot and must not decide about the framework within which it operates.


Daher ist eine Strategieüberprüfung inner- und ausserhalb der SNB sinnvoll. Geldpolitisch prüfenswert sind das Inflationszielband, die Zentralität des Zinsinstruments und die Kommunikation. Die Glaubwürdigkeit der SNB verbietet ein Auseinanderklaffen von Theorie und Praxis, aber auch allzu häufiges und detailversessenes Feilen an der Strategie, und sie verlangt Konzentration auf das Wesentliche. Gleichzeitig sollte die SNB ihre Bindung an den – gegebenenfalls sich wandelnden – Willen des Gesetzgebers betonen. Bei Fragen, die nicht allein in ihre Zuständigkeit fallen, muss sie klarstellen, dass sie Partei und nicht Schiedsrichterin ist. Damit die SNB auch in Zukunft zu den grossen Schweizer Erfolgsgeschichten zählt, muss sie von Zeit zu Zeit über die Bücher gehen. Doch alleine kann sie die Verantwortung in Geld- und Währungsfragen nicht tragen.

German Banks Send Mixed Signals on Digital Euro

In the FAZ, Christian Siedenbiedel reports that Deutsche Bank questions whether a digital Euro as envisioned by the ECB (i.e., with tight quantity restrictions) would be successful:

Die Argumentation geht so: Die EZB will den digitalen Euro einführen, um auf den verstärkten Währungswettbewerb zu antworten. … Um sich vor solchem Machtverlust sowohl durch Digitalgeld von anderen Notenbanken („Krypto-Dollars“) als auch durch privates Digitalgeld („Global Stable Coins“) zu schützen, treibe die EZB den Digitaleuro voran. Also aus längerfristigen politischen Motiven. Dabei sei unklar, ob der digitale Euro sich international am Markt durchsetzen könne und ob die Menschen in der Eurozone dafür überhaupt Bedarf hätten. “Das Design des digitalen Euros, soweit bisher bekannt, lässt erwarten, dass die potentiellen Nutzer kaum einen Unterschied zu bestehenden Bezahloptionen erkennen werden”.

Update: From the dbresearch document prepared by Heike Mai:

Lifting the limits on how much each user can hold would change the situation entirely, allowing a massive outflow of bank deposits into the digital euro. As a result, lending decisions and money creation would shift from the decentralised, privately owned banking sector to a central, state-run authority: the ECB. In this case, Europe would face the fundamental question of which type of monetary and financial system it wants. The answer to that would have to come from democratically elected representatives.

The German Banking Industry Committee sees a central role for the digital Euro, however, according to a new paper:

In a policy paper, the German Banking Industry Committee (GBIC) for the first time sets out detailed thoughts on the design of a “digital euro”. In this paper, experts from Germany’s five national banking associations draw up an ecosystem of innovative forms of money that extends far beyond the idea of digitalised central bank money, which is referred to as Central Bank Digital Currency (CBDC). The ECB will probably launch the project for a digital euro in mid-July 2021.

“To be successful, the digital euro must do three things: It must be as easy for consumers to handle as cash. It must be viable in the long term for business enterprises, e.g. for automated machine-to-machine payments. And the digital euro must be well embedded in our delicately balanced, carefully secured and highly regulated European financial system because this system guarantees safe and fair access to financial and banking services for everyone in Europe”, notes Dr Joachim Schmalzl, executive member of the Board of Management of the German Savings Bank Association (DSGV), which is currently the lead coordinator for the German Banking Industry Committee.

In the opinion of the experts from Germany’s five national banking associations, issuing money should remain the responsibility of credit institutions in the proven two-tier banking system [my emphasis], even if the digital euro becomes legal tender like cash. For this reason, the ecosystem of digital money which they propose is made up of three key elements:

  • retail CBDC for private use
  • wholesale CBDC for commercial and savings banks
  • tokenised commercial bank money for use in industry

Retail CBDC issued by the central bank is to be used by private individuals in the euro area in the same way as cash for everyday payments, e.g. to retailers or government agencies. It should be possible to use the digital euro like cash, anonymously and offline. For this purpose, credit institutions will provide consumers in Europe with “CBDC wallets”, i.e. electronic wallets.

Wholesale CBDC issued by the central bank is to be used for the capital markets and interbank transfers. The GBIC’s experts are calling for this special form of the digital euro partly because, by adopting this approach, the ECB would be able to include further digitalisation of central bank accounts in its project. The ultimate aim is to achieve improvements which can benefit consumers, enterprises and also the banking sector.

Tokenised commercial bank money, which will be made available by commercial and savings banks, is to complement the two forms of digital central bank money, in particular to meet corporate demand arising from Industry 4.0 and the Internet of Things. Tokenised commercial bank money could facilitate transactions based on “smart” – i.e. automated – contracts and thus increase process efficiency.

“Increasing process digitalisation and automation will provide completely new opportunities for Europe’s enterprises. The banking sector is ready to provide new solutions for its corporate customers by issuing innovative forms of money. The ECB must define the necessary framework that will enable Europe’s banking sector and real economy to make reasonable use of the new opportunities”, Joachim Schmalzl observed on behalf of the GBIC.

I share the skepticism of DB research. And I can understand that banks prefer to maintain the two-tiered system while pushing for broader and more efficient payment options for their business clients.

“Schuldenbremse — Licht und Schatten (Debt Brake—Merits and Risks),” FuW, 2021

Finanz und Wirtschaft, June 5, 2021. PDF.

  • The debt brake addresses some political economy frictions, but not others.
  • Focusing too narrowly on explicit government debt it provides incentives to accumulate implicit debt, sell assets, or engage in creative accounting.
  • The political pressure to raise SNB profit disbursements is a symptom of these incentives.

“Die Schattenseiten von Schuldenbremsen (The Dark Side of Debt Limits),” ifoSD, 2021

ifo Schnelldienst 4/2021, April 14, 2021. PDF.

Was Schuldengrenzen aus politökonomischer Sicht besonders attraktiv erscheinen lässt – ihre vermeintliche Einfachheit und Klarheit – birgt also auch Risiken. Es führt dazu, dass Politiker und ihre Wähler die Solidität der Staatsfinanzen über Gebühr an expliziten Bruttoschulden messen. Was aber zählt, wenn es um unerwünschte Umverteilung zulasten künftiger Generationen geht, ist staatliches Nettovermögen in einer umfassenden Gesamtschau.

“Money Creation, Bank Profits, and CBDC,” VoxEU, 2021

VoxEU, February 5, 2021. HTML.

Based on CEPR DP 15457, I assess possible implications of the introduction of retail CBDC for bank profits. The model implies annual implicit subsidies to U.S. banks of up to 0.8 percent of GDP during the period 1999-2017.

SNB Profit Distributions

The Federal Department of Finance and the SNB have agreed on a new scheme for the distribution of SNB profits. Agreement for the period 2020-2025, Explanations. Some comments in German (also available as PDF):

Profitieren Bund und Kantone finanziell von den höheren SNB-Ausschüttungen?

  • Höhere Gewinnausschüttungen in der Gegenwart bedingen tiefere in der Zukunft.
  • In erster Näherung bleibt das Nettovermögen von Bund und Kantonen unverändert, denn es berücksichtigt auch den Wert der zukünftigen Ansprüche gegenüber der SNB.
  • Siehe z.B. „Die Volkswirtschaft“ 8-9 2020, HTML.

Warum dann die positiven Reaktionen bei Vertretern von Bund und Kantonen?

  • Politiker/Wähler orientieren sich an den ausgewiesenen Schulden des Staates. Höhere Ausschüttungen ermöglichen eine tiefere Schuldenaufnahme. Daher die Reaktionen.
  • Relevanter als ausgewiesene Schulden ist das Nettovermögen. Dieses wird von Ausschüttungen (in erster Näherung) nicht beeinflusst.

Was ist die primäre Wirkung höherer Ausschüttungen?

  • Die Schuldenbremse wird in der Gegenwart gelockert und in der Zukunft angezogen.
  • Falls die Schuldenbremse bindet, erhöhen frühere, höhere Ausschüttungen den Spielraum für staatliche Ausgaben in der Gegenwart, aber nicht in der Zukunft.
  • Hohe Ausschüttungen könnten über ihre Wirkung auf das Eigenkapital der SNB auch deren geldpolitische Entscheide beeinflussen.

Wie ist die neue Gewinnausschüttungsformel ökonomisch zu bewerten?

  • Ausschüttungen sollten die Geldpolitik nicht konterkarieren.
  • Die Geldpolitik setzt Bilanzlänge und -struktur der SNB als Instrumente ein. Ihre Glaubwürdigkeit kann vom Eigenkapital der SNB abhängen.
  • Demnach müssten Ausschüttungen von Bilanzlänge, -struktur und Eigenkapital der SNB abhängen. Nicht vom Gewinn des Vorjahres.

Wie ist die neue Gewinnausschüttungsformel politökonomisch zu bewerten?

  • Problematisch ist, dass der Eindruck entstehen kann, alle paar Jahre würde unter politischem Druck um eine neue Formel gefeilscht. Die Bindung an eine sinnvolle Regel würde diesem Eindruck entgegenwirken.
  • Gleichzeitig reduziert die neue Formel den politischen Druck. Sie signalisiert die Bereitschaft der SNB zur Diskussion.

Wie könnte ein alternatives Ausschüttungsmodell aussehen?

  • Die SNB erklärt periodisch, welche Bilanzlänge und -struktur sie zur Erfüllung ihrer Aufgaben benötigt. Sie stellt sich der Kritik, entscheidet aber eigenverantwortlich.
  • Ausschüttungen sind nicht zweckgebunden. Dadurch wird vermieden, dass sich Interessengruppen bilden, die systematisch auf höhere Ausschüttungen drängen.

Welches Grundproblem bliebe bestehen?

  • Eine ökonomisch begründete Ausschüttungspolitik führt zu fluktuierenden Ausschüttungen. Diese können die Schuldenbremse konterkarieren.
  • Glättet die SNB hingegen ohne geldpolitische Notwendigkeit ihre Ausschüttungen, masst sie sich eine Kontrolle der Fiskalpolitik an, die ausserhalb ihres Aufgabenbereichs liegt.

Comments on Geneva Report 23

Panel with Elga Bartsch, Agnès Bénassy-Quéré, Giancarlo Corsetti, Olivier Garnier, and Charles Wyplosz. Moderated by Tobias Broer.

Elga Bartsch, Agnès Bénassy-Quéré, Giancarlo Corsetti, Xavier Debrun: Geneva Report 23 | It’s All in the Mix: How Monetary and Fiscal policies Can Work or Fail Together.

Event at PSE.

My comments on the report.

“Staatsschulden sind keineswegs kostenlos (Free Government Debt?),” NZZ, 2021

NZZ, February 1, 2021. PDF (title changed by NZZ). Related article in Ökonomenstimme. HTML.

Do negative interest rates render government debt costless? No. What about r<g? I discuss Olivier Blanchard’s presidential address and the conclusions that columnists have drawn.

For background: See this post.

“The Pandemic Endgame,” VoxEU, 2021

VoxEU, January 11, 2021, with Martin Gonzalez-Eiras. HTML.

Based on the CEPR discussion paper, we draw conclusions for the pandemic endgame. We explain why Israel will likely impose a harsher lockdown than other countries, especially poor ones. And why we should expect “inverse lockdowns”—measures to stimulate social interaction.

John Cochrane about CBDC and Me

Writing about CBDC, John Cochrane makes it clear that he is in favor. He links to my work and writes

Dirk Niepelt has written a lot about CBDC theory, including reserves for all in 2015, a recent Vox-EU summary and papers,  here with Markus Brunnermeier a JME paper “CBDC coupled with central bank pass-through funding need not imply a credit crunch nor undermine financial stability,” a follow up including “The model implies annual implicit subsidies to U.S. banks of up to 0.8 percent of GDP during the period 1999-2017.”  Here  “reserves for all” “does not affect macroeconomic outcomes,”

Not Much Left of “Modern Monetary Theory”

Alberto Bisin (Journal of Economic Literature, December 2020) reviews Stephanie Kelton’s “The Deficit Myth: Modern Monetary Theory and the Birth of the People’s Economy:”

Never is its logical structure expressed in a direct, clear way, from head to toe. … Some of these statements are literally correct but used for incorrect or misleading implications—plays on words, effectively. They seem taken directly from the book of tricks of the Greek sophists (the ones Aristophanes makes fun of).

John Cochrane (blog post, July 2020) reviews the same book:

Skeptics have called it “magical monetary theory.” They’re right.

Dirk Niepelt (blog post/Neue Zürcher Zeitung (in German), April 2019):

The Macroeconomic Perpetuum Mobile.

“Dirk Niepelt imäch (Interview with Dirk Niepelt),” swissinfo, 2020

Swissinfo, December 14, 2020. HTML, podcast.

We talk about CBDC, the Swiss National Bank, whether CBDC would render it easier to implement helicopter drops, and how central bank profits should be distributed.

“Optimally Controlling an Epidemic,” CEPR, 2020

CEPR Discussion Paper 15541, December 2020, with Martin Gonzalez-Eiras. PDF (local copy).

We propose a flexible model of infectious dynamics with a single endogenous state variable and economic choices. We characterize equilibrium, optimal outcomes, static and dynamic externalities, and prove the following: (i) A lockdown generically is followed by policies to stimulate activity. (ii) Re-infection risk lowers the activity level chosen by the government early on and, for small static externalities, implies too cautious equilibrium steady-state activity. (iii) When a cure arrives deterministically, optimal policy is dis-continous, featuring a light/strict lockdown when the arrival date exceeds/falls short of a specific value. Calibrated to the ongoing COVID-19 pandemic the baseline model and a battery of robustness checks and extensions imply (iv) lockdowns for 3-4 months, with activity reductions by 25-40 percent, and (v) substantial welfare gains from optimal policy unless the government lacks instruments to stimulate activity after a lockdown.

“Wirtschaftspolitik in Corona-Zeiten (Economic Policy in Times of Corona),” FuW, 2020

Finanz und Wirtschaft, December 9, 2020. PDF.

  • Economic policy is not about GDP growth. It’s about welfare.
  • Externalities are key. Infection externalities don’t go away by calling for responsible behavior. Infection externalities can turn positive.
  • Keeping worthy companies or networks alive does not require government intervention, unless capital markets don’t work.
  • To judge the right amount of burden sharing is beyond economics. But economics gives some clues: In an ideal world, idiosyncratic risk exposure would be insured while in second best, taxes and subsidies achieve only part of that. The data show that trade-offs between public health and economic activity are less severe than sometimes argued.

The Economist on CBDC and Disintermediation

The Economist discusses the risk of CBDC-induced bank disintermediation. Their summary of the 2019 paper by Markus Brunnermeier and myself:

If people prefer CBDCS, however, the central bank could in effect pass their funds on to banks by lending to them at its policy interest rate. “The issuance of CBDC would simply render the central bank’s implicit lender-of-last-resort guarantee explicit,” wrote Markus Brunnermeier of Princeton University and Dirk Niepelt of Study Centre Gerzensee in a paper in 2019. Explicit and, perhaps, in constant use.

“CBDC: State of Play, Practical Challenges, Open Issues,” SUERF Webinar, 2020

SUERF Webinar “CBDC: State of play, practical challenges, open issues” with Ulrich Bindseil (ECB) and Morten Bech (BIS). Moderated by Dirk Niepelt. December 4, 2020, 2 pm.

“Monetary Policy with Reserves and CBDC: Optimality, Equivalence, and Politics,” CEPR, 2020

CEPR Discussion Paper 15457, November 2020. PDF (local copy).

We analyze policy in a two-tiered monetary system. Noncompetitive banks issue deposits while the central bank issues reserves and a retail CBDC. Monies differ with respect to operating costs and liquidity. We map the framework into a baseline business cycle model with “pseudo wedges” and derive optimal policy rules: Spreads satisfy modified Friedman rules and deposits must be taxed or subsidized. We generalize the Brunnermeier and Niepelt (2019) result on the macro irrelevance of CBDC but show that a deposit based payment system requires higher taxes. The model implies annual implicit subsidies to U.S. banks of up to 0.8 percent of GDP during the period 1999-2017.

“Unabhängigkeit der Nationalbank (Independence of the SNB),” FuW, 2020

Finanz und Wirtschaft, July 25, 2020. PDF.

The Swiss National Bank—yes, the Swiss one—feels it must remind politicians of its independence. Parliamentarians from left to right (!) voice demands. To shrink the SNB’s balance sheet? No, for more central bank profits to be distributed sooner rather than later.

I discuss misconceptions, possible motivations, and a constructive response. «The best way to defend the independence of a central bank is never to exercise it.»