Sunday, February 2, 2014

Who signs a country's banknotes?

2010 Bank of England note signed by Andrew Bailey, former Chief Cashier of the Bank.

A few years ago, Peter Stella and Åke Lönnberg conducted a study that classified national banknotes by the signatories on that note's face. They found some interesting results. Of the world's 177 banknotes with signatures (10 had no signature whatsoever), the majority (119) were signed by central bank officials only. Just four countries issue notes upon which the sole signature was that of an official in the finance ministry: Singapore, Bhutan, Samoa, and (drum roll) the United States.

Stella and Lönnberg hypothesize that the signature(s) on a banknote indicate the degree to which the issuing central bank's is financially integrated with its government. The lack of a signature from a nation's finance ministry might be a symbol of a more independent relationship between the two, the central bank's balance sheet being somewhat hived off from the government's balance sheet and vice versa. The presence of a finance minister's signature would indicate the reverse, that both the treasury and central bank's balance sheets might be best thought of as one amalgamated entity.

The nature of this arrangement is significant because if something disastrous were to happen to an independent central bank's financial health, say its assets were destroyed and all hope of profits dashed for eternity, the central banker should not necessarily expect support from his/her government. Lacking in resources, monetary policy could go off the rails. (Why would it go off the rails? Here I go into more detail).

On the other hand, should it be established by law that a government is to backstop its central bank, that same disaster would pose a smaller threat to monetary policy since the nation's finance minister, his John Hancock affixed to the nation's notes, would presumably come to the central bank's rescue.

These ideas are similar to Chris Sims's classification of type F and type E central banks (alternative link). One of the features of type F banks (like the Fed) is that "there is no doubt that potential central bank balance sheet problems are nothing more than a type of fiscal liability for the treasury." On the other hand, with type E banks (like the ECB) "it is not obvious that a treasury would automatically see central bank balance sheet problems as its own liability."

So is it the case that the Federal Reserve is actually more fused with the U.S. Treasury than other central banks are? One reading of the Federal Reserve Act might indicate yes. Section 16.1 stipulates that Federal Reserve notes are ultimately obligations of the US government:
Federal reserve notes, to be issued at the discretion of the Board of Governors of the Federal Reserve System for the purpose of making advances to Federal reserve banks through the Federal reserve agents as hereinafter set forth and for no other purpose, are hereby authorized. The said notes shall be obligations of the United States and shall be receivable by all national and member banks and Federal reserve banks and for all taxes, customs, and other public dues.
The language in the above phrase would seem to indicate that should the Fed find itself incapable of exercising monetary policy (Stella and Lönnberg  use the term policy insolvency), the US government is obliged to step in and make good on the Fed's promises, however those promises might be construed. The fact that Treasury Secretary Jacob Lew's signature appears on all paper notes, as does that of U.S. Treasurer Rosa Gumataotao Rios, can perhaps be taken as an indication of this guarantee.

Bank of Canada notes, on the other hand, are signed by the Governor of the BoC and his deputy. Finance Minister Flaherty's signature is nowhere in sight. This jives well with a quick reading of the Bank of Canada Act, which stipulates that though notes are a first claim on the assets of the Bank of Canada, the government itself accepts no ultimate obligation to make good on banknotes. In theory, should the Bank of Canada cease to earn a profit from now to eternity, Canadian monetary policy could go haywire—American monetary policy, backstopped by the Treasury, less so.

Other central banks go even further in formalizing this separation. In Lithuania, for instance, the law states that: "The State of Lithuania shall not be liable for the obligations of the Bank of Lithuania, and the Bank of Lithuania shall not be liable for the obligations of the State of Lithuania." Should Leituvos Bankas hit a rough patch, so will its monetary policy.

Stella and Lönnberg correlate the rise of independent central banking with a movement away from the printing of finance minister signatures on notes. For instance, the sole signature on Euro banknotes is that of the President of the ECB, Mario Draghi. Two of the currencies replaced by the Euro, the Irish punt and the Luxembourg franc, which had carried signatures of finance department officials, no longer exist, symbolic evidence of the Euro project's dedication to central bank independence.

Sims uses the ECB as an exemplar of type E central banks because "the very fact that there is a host of fiscal authorities that would have to coordinate in order to provide backup were the ECB to develop balance sheet problems suggests that such backup is at least more uncertain than in the US." For evidence, he points to the fact that the Fed carries just 1.9% of its balance sheet in capital and reserves while the ECB holds 6.7%.

Stella and Lönnberg hint at the prevalence of a "rather singular U.S. view of central bank and treasury relations." My interpretation of this is that most conversations about central banking are inherently conversations about the the world's dominant monetary superpower, the Federal Reserve. This is surely evident in the blogosphere, where we mostly talk as-if we were Bernanke, not Carney or Poloz or Ingves (Lars Christensen is a rare counter-example who is fluent in multiple "languages"). In the same way that all Americans only understand English while all foreigners are conversant in English and their native tongue, non-American commentators like me can't talk solely in terms of our own central bank (in my case the Bank of Canada) lest we fall out of the conversation. The Fed becomes our focal point.

Yet among central banks, the Fed is an odd duck, since the wording in the Federal Reserve Act and the signature on its notes would indicate a more well-integrated financial relationship between central bank and treasury than most. The upshot is that popular conceptions of the central banking nexus will often be wrong as they will be couched in terms of the U.S.'s integrated viewpoint, whereas most of the world's central banks are not structured in the same way as the U.S. A deterioration of the Fed's balance sheet would likely be neutral with respect to monetary policy, but for many of the world's nations this simply isn't the case.

On a totally unrelated side note, I found it interesting that Bank of England notes stand out as being signed by the Chief cashier of the Bank, not the governor. When the BoE opened its doors for business in 1694, the banknotes it issued were written on blank sheets of paper, often for unusual quantities (standardized round numbers were not introduced till the 1700s). The bank's directors and its governor, usually well-established bankers who simultaneously ran their family business, were not responsible for the BoE's day-to-day operations, this being devolved to the bank's cashiers who were given the repetitive task of signing each note by hand. Even when the ability to print signatures directly on to notes was developed in the 1800s, the practice of affixing the cashier's signature continued, despite the fact that mechanical process would make it easy for the higher-ranked governor to get his name on each note.

So while Mark Carney's route from the BoC to the BoE got him a higher salary, more prestige, and posher digs, in one respect his standing has deteriorated: there are no longer millions of bits of paper circulating with his name on them. Chief cashier Chris Salmon has that distinction.

PS: You should try and read all of Peter Stella's papers. They are excellent. He blogs here.


  1. Nice post and nice reference.

    Btw, the best answer to "I promise to pay" is given by the Reserve Bank of India:

    What is the meaning of "I promise to pay" clause?

    As per Section 26 of Reserve Bank of India Act, 1934, the Bank is liable to pay the value of banknote. This is payable on demand by RBI, being the issuer. The Bank's obligation to pay the value of banknote does not arise out of a contract but out of statutory provisions.

    The promissory clause printed on the banknotes i.e., "I promise to pay the bearer an amount of X" is a statement which means that the banknote is a legal tender for X amount. The obligation on the part of the Bank is to exchange a banknote for coins of an equivalent amount.

    1. Neat! One weird thing about Canada (I wonder if any other countries are the same?) is that the Bank of Canada issues notes and the Royal Mint issues coins, and the two are quite distinct. (I'm not sure of the exact legal and policy distinction involved.)

    2. Ramanan, who signs Indian notes?

      Nick, pretty sure the U.S. and U.K. do the same thing. Mints seem to have been always controlled by the sovereign, central banks were private (at least initially so). Perhaps the modern separation is an echo of this historical split?

    3. This comment has been removed by the author.

    4. JP,

      The Reserve Bank Governor signs it.

      Funnily, India used to have a one-rupee note signed by the Secretary, Ministry of Finance (in addition to issuing coins) with others signed by the Reserve Bank Governor but it no longer issues those one-rupee notes.

    5. Finally something I can contribute to:

      "pretty sure the U.S. and U.K. do the same thing."

      Yes, the US does. The Mint (part of Tsy) mints the coins and sells them at face value to the Fed (and it loses money on pennies and nickles!). Whereas notes are printed by the Bureau of Engraving and Printing (also part of Tsy), but they are sold at production cost to the Fed.

      Coins at the Fed are Fed assets. Notes at the Fed are not on the balance sheet, but only go on the balance sheet at their face value once they are sold to banks.

      I'm not 100% on all that, but I have looked into it and that was my conclusion. I might even be able to dig up references on that.

    6. "Notes at the Fed are not on the balance sheet, but only go on the balance sheet at their face value once they are sold to banks."

      ... and of course they go on as liabilities.

  2. JP: "In the same way that all Americans only understand English while all foreigners are conversant in English and their native tongue, non-American commentators like me can't talk solely in terms of our own central bank (in my case the Bank of Canada) lest we fall out of the conversation. The Fed becomes our focal point."

    Yep. I make a point of writing "suppose the central bank....." or "suppose the Bank of Canada......" and commenters keep responding "But if the Fed...."

    1. Their loss. It's good to speak two languages -- it gives you perspective.

  3. Good post and unexpected angle, since I usually hear about governments plundering their central banks, rather than backing them up.

    1. Mike, I kid you not. Here's an example from Stella/Lonnberg:

      "The negative equity problem of the Central Bank of Chile can be traced back to the financial crisis of 1982 the cost of which has been estimated at 33 percent of GDP. Initially the cost was borne entirely by the central bank but in 1987 the government made a partial recapitalization. This issue was not faced again until 2006 when the Congress approved a transfer of up to 0.5 percent of GDP per annum for a period of 5 years. The central bank has estimated that this capital infusion would advance the date at which it would have positive equity from 2041 to 2026, that is, by fifteen years. The central bank received US$606 million as the first part of this partial recapitalization on December 27, 2006."