Saturday, November 30, 2013

The three lives of Japanese military pesos

1942 Japanese Invasion Philippines Peso with a JAPWANCAP Stamp

In his reply to Mike Sproul, Kurt Schuler brings up the question of the determination of the value of a very peculiar kind of money: military currency. Curious, I investigated one example of such money, Japanese-issued "invasion money" in Philippines both during and after World War II. As best I can tell, the mechanism by which the value of these military notes has been mediated has gone through three different phases, each of them teaching us something interesting about money.

In January 3, 1942, a few weeks after successfully invading Philippines, the Japanese Commander-in-Chief announced that occupying forces would henceforth use military-issued currency as legal tender. Notes were to circulate at par with existing Philippines "Commonwealth" Pesos. Since this military scrip was not directly convertible into existing pesos, the trick to get it to circulate at par can probably be found in the tersely titled proclamation Acts punishable by death which, among seventeen acts that could result in loss of life, listed the thirteenth as:
(13) Any person who counterfeits military notes; refuses to accept them or in any way hinders the free circulation of military notes by slanderous or seditious utterances.
This is a great example of a Warren Mosler fiat money. According to Mosler, the state's requirement that citizens discharge their tax obligation with a certain intrinsically worthless medium on pain of being shot in the head is sufficient to give that medium a positive value. Likewise, requiring citizens to use the same medium in the course of regular payments and accept it to discharge debts, all on pain of death, would probably have promoted a positive value for intrinsically worthless paper.

Over time, those bits of "forced" paper will enjoy constant purchasing power as long as the issuer withdraws excess currency or adds it when desired. This is the quantity theory of money.

By 1943, however, it seems that the Japanese occupying forces, now being pushed back by the Allied forces, were desperately issuing excess notes to pay for operations. The Filipino monetary system proceeded to run smack dab into Gresham's law. The unit of account, the peso, was defined in terms of two different media—original pesos and military pesos. This meant that debtors could choose to discharge their debts with either. However, if one was perceived to be more valuable than the other, this superior medium would be hoarded and the inferior one used to pay off the debt. Legacy pesos had completely disappeared from circulation by 1943—only war pesos were being used to discharge debts and pay for goods, a decent indicator that the value of Japanese invasion pesos had fallen below that of original pesos. Bad money had chased out the good. (See [1] and [2] for evidence of Gresham's law)

Through 1944 and 1945, the war peso would endure extreme inflation. 10P had been the largest denomination in 1942. The military introduced 100P, 500P, and 1000P notes in subsequent years. In Neil Stephenson's Cryptonomicon, a wide-ranging historical/science fiction novel filled with monetary themes, there's an interesting passage in which Japanese soldier Goto Dengo describes the use of military scrip, probably sometime in 1943 or 1944:
The owner comes over and hands Goto Dengo a pack of Lucky Strikes and a book of matches. "How much?" says Goto Dengo, and takes out an envelope of money that he found in his pocket this morning. He takes the bills out and looks at them: each is printed in English with the words THE JAPANESE GOVERNMENT and then some number of pesos. There is a picture of a fat obelisk in the middle, a monument to Jose P. Rizal that stands near the Manila Hotel.
The proprietor grimaces. "You have silver?"
"Silver? Silver metal?"
"Yes," the driver says.
"Is that what people use?" The driver nods.
"This is no good?" Goto Dengo holds up the crisp, perfect bills.
The owner takes the envelope from Goto Dengo’s hand and counts out a few of the largest denomination of bills, pockets them, and leaves.
Goto Dengo breaks the seal on the pack of Lucky Strikes, raps the pack on the tabletop a few times, and opens the lid.
Japanese invasion currency, already being well on its way to being repudiated, would become completely worthless upon Japan's unconditional surrender in 1945.

Well, not entirely valueless. The second chapter in the life of military scrip begins with The Japanese War Notes Claimants Association of the Philippines, or JAPWANCAP. Formed in 1953 on behalf of Filipinos left holding stranded quantities of worthless Japanese invasion money, JAPWANCAP's mission was to hold both the US and Japanese government's liable for the redemption of war currency (the US had also issued counterfeit Japanese military currency). So while pesos had been valued prior to the war's end upon pain of death, and their value regulated by limiting the quantity outstanding, those same pesos were now valued on the margin as a liability of their issuer. Given the possibility of redemption, an old invasion note was worth more than zero.

Was JAPWANCAP successful? While the case was heard in a United States Court of Claims in 1967, it was thrown out on a technicality, the statute of limitations having had passed. Put simply, the court would not hear a claim that had not been filed within six years of that claim first being accrued, and in JAPWANCAP's case many more years than that had already passed.

This makes one wonder, if Filipinos in 1953 were already convinced that Japanese invasion pesos were the liability of the issuer, and therefore redeemable in some quantity of yen or dollars, did that same motivation also lead them to originally accept new military pesos in 1942? To what degree was the initial acceptance of pesos driven by the threat of force (& subsequent changes in value regulated by their quantity), and to what degree was their value dictated by their status as a liability of a well-backed issuer? That's a question we can never be entirely sure of. But while the force/quantity theory story fits the facts, the liability story does too. The military peso's inflation, for instance, can be attributed to the rising quantity of money, but also to the increasing likelihood of Japan losing the war, a loser's liability's being worth far less than a winner's.

Which brings us to the third chapter in the evolution of Japanese military pesos. Nowadays you can buy the notes on eBay for a few bucks. Their value is no longer dictated by gunpoint, nor by their liability nature, but their existence as a unique commodity, much like gold, bitcoin, or some rare antique.

To learn more, here is a paper called "Financing Japan’s World War II occupation of Southeast Asia"

Note: I will be posting sparsely over the next two months, probably once every two weeks.


  1. One more complication: Death threats are a form of backing, and the ability to kill people is an asset. As long as the money issuer has the ability to kill me, I'll value its currency. But if the issuer loses the ability to kill me, I will stop valuing its currency just like in the case of a money-issuing bank that had lost the ability to pay me silver for its money.

    1. Dunno Mike, that's a pretty broad definition of backing. Just because a producer of, say, pink pebbles threatens to kill anyone upon non-acceptance of those pebbles as exchange media doesn't mean that those pebbles are a financial liability of the producer.

    2. I'll have to give you that one. There's backing, and there's theft, and there's only limited overlap between the two. If the government says "A must accept an occupation peso from B as if it were a real peso." then that's theft. If the government says "The tax man will accept occupation pesos as if they were real pesos, but if you don't pay taxes then we'll kill you." then that's backing.

  2. Great post again!
    My impression is that a lot of the reason why people accept money is because they WANT to sell excess that is not usable by themselves. That is especially true when the excess is man-hours (or machine-hours) that if left unsold would entirely go to waste anyway. Almost always, the real economy is mostly demand constrained. So basically a money issuing authority is pushing on an open door when forcing acceptance of money on a population. The only challenge comes from alternative types of money BUT as you say, Gresham's law shows that even that isn't much of a challenge as hoarding soon takes "good" money out of circulation and leaves the economy thirsting for circulating money.

    1. Let's not white wash what the Japanese did. They basically forced Filipinos to surrender useful goods for paper when they would otherwise have preferred pesos or dollars. It was theft. Luckily, forced currencies that involve gun-point coercion are very rare. I can only find examples during wartime.

    2. You are quite correct and I certainly don't mean to underplay how despicable the Japanese invasion was.
      My (badly made) point though was that perhaps getting acceptance of the currency wasn't that tough a job. Your previous post about use of "orphan" currencies in Somalia and Iraq shows that currencies with nothing backing them do sometimes get used even when no threat of force exists doesn't it?

      I guess another (equally repugnant) example of forced currency was the "hut tax token" system used in colonial Africa. The only way to obtain hut tax tokens was by collaboration with the colonial authorities or by trading with someone who had collaborated. I don't know whether you would count that as a formal war. It was a hostile exploitative occupation.

      Perhaps part of the reason why gun point coercion is only needed in times of resented occupation is because people then refuse money mainly because they don't want to be complicit with the occupiers?

    3. Both the invasion itself and the use of forced currency were despicable. The Somalia and Iraq stories are about an existing currency that citizens continued to use by choice after the original issuer ceased to exist. Japanese pesos, on the other hand, were a new currency, not an existing currency, forced upon Filipinos rather than chosen by them. Very different situations.

  3. I guess even popular, democratic, governments have the threat of imprisonment as an ultimate sanction to cause tax demands to be paid. That isn't gun-point coercion but it is coercion forcing use of the currency isn't it?

    1. No, modern governments don't run forced currency systems. Legal tender laws are non-binding on anyone who wishes to conduct their affairs in a foreign currency. A US shopkeeper need only hang a sign in his shop door saying "Canadian bank notes accepted only", and negotiate with his creditors/debtors that all debts must be discharged with Canadian paper money. See my legal tender posts. People choose currency, it isn't forced on them.

    2. I agree that currency isn't forced on people, but as Larry White's talk on the Liberty dollar and E-gold shows, governments can still harass alternative currencies out of existence.

      Aside from overt govt suppression, I think the capital gains tax is the prime obstacle to alternative monetary arrangements (e.g. the IRS treats exchanges of gold coins for goods as barter, subject to cap. gains; who wants to deal with the record keeping hassle and/or prospect of paying taxes on transactions? Far better to stick with USD). Kevin Dowd speculated that financial assets might play a greater role in interbank clearing in a mature free banking system, but I don't see how that would be possible due to capital gains taxes.

    3. Yes, I agree about the capital gains tax. Spending away bitcoins is a pain if you have to calculate the gain you've made on each transaction. Ditto for capital losses if you want to claim them.

  4. JP Koning, thanks for the pointer to your earlier posts. This is very interesting. Please bear with me because I am coming from the starting point of having been "taken in" by the meme that modern currencies have a largely chartalist basis and I'm having to get my head around your refutation of that whole idea.
    I'm struggling with your point about taxes being payable with non-legal tender such as cheques or Scottish bank notes (in the UK). Is it not true that although as a tax payer I may personally hand over a cheque BUT as a consequence of that my bank must get hold of bank reserves and transfer those bank reserves to the government? So the ultimate result is that demand for the bank reserves is forced by the taxation? Tax payment must be in a form that entails my bank transferring GBP bank reserves to the government?
    An anecdote that may be relevant is that when we were on holiday in Croatia, the small Croatian holiday company were very apologetic in saying that we had to make any debit card payments to them in Croatian Kuna even though prices had been advertised in Euro. They said that they were not legally allowed to transact in Euro with debit cards. We didn't mind at all because for us as UK citizens Euro no easier than Kuna anyway. I did think this was all quite curious.
    My other half is self employed in the UK but has more than half of her customers pay in either USD or Euros because they are foreign. She also previously worked for a UK start up that was purchased by a German company and yet the takeover was paid for in USD and she got a (small) part of the takeover payout that came in as USD to her. None of this had shaken my acceptance of chartalist ideas because I thought our bank exchanged the FOREX before it got to our bank account and we paid tax in GBP.

    1. " So the ultimate result is that demand for the bank reserves is forced by the taxation?"

      No, that's a non sequitur. The observation that taxes are ultimately settled in central bank reserves does not mean that taxes drive the use or value of those reserves. Taxes, like all transactions, happen to be settled with central bank money because the clearinghouse function is monopolized by the central bank. If the government decided that it would henceforth only accept taxes in physical gold, any unwanted reserves would be repurchased by the central bank at a rate sufficient to hit its inflation targets, and remaining reserves would continue to circulate as a popular form of exchange media for non-government payments. It would be the same if Walmart announced it would only accept gold. Reserves would continue to be valued and circulate as exchange media in non-Walmart payments.

      The fact that Somali shillings were still used and valued despite the non-existence of taxes should put to rest the odd idea that fiat money requires tax payment to maintain its positive value.

    2. For the thought experiment example your giving, would the government be paying state employees with central bank money and receiving taxes in the form of gold bullion? Would the government exchange gold bullion for central bank money so as to obtain the central bank money for making the payments?

    3. Either/or.

      Anyways, I don't want to turn this into a big conversation about tax-driven money since this post was ultimately about legal tender-driven money in a war setting -- the Japanese occupiers did not levy taxes. For my views on chartalism, try these posts:

    4. Hi stone, for an interesting take on the topic I recommend "The tax-foundation theory of fiat money" and "Famous Myths of Fiat Money" by Dror Goldberg (and thanks to JP for making me aware of Goldberg's work).

    5. Thanks Peter and JP for all the links and info

  5. Quick off-topic comment.

    You probably saw this already, but some BTC tracking websites have switched to mBTC (no exchanges have yet, to my knowledge). I think this will be rather beneficial for the adoption of bitcoin. Upon seeing the chart, my immediate reaction was, "Wow, this looks like a real currency now." Obviously, nothing changed except for decimal point placement, but there's something psychologically reassuring about seeing 1 mBTC = $1.15 rather than 1 BTC = $1150. In short, it makes it look less bubbly and more like a legitimate currency. (I think there would be a psychologically damping effect on swings too; $1.15 to 0.85 just feels less crazy than $1150 to $850, even though there's no difference).

    As you have noted, units of account need to be stable, and BTC doesn't fit the bill now (perhaps never). But I still think using mBTCs can at least start the process of getting people to start thinking in terms of BTC , and that's important for it to be used as MOE. Sites like bitcoinstore still price in bitcoins, and seeing prices like "0.256 BTC" is just frankly weird. Using mBTC will give people a mental shortcut--"Oh, a MilliBit, that's a bit more than a dollar (or Euro)." That's a big improvement in usability.

    I think the importance of making bitcoin prices more familiar and relatable to dollar prices will be important, even if it's essentially a fictional adjustment.

  6. Totally off-topic, JPK, but I think you will find this BBC radio history of the dollar in five fifteen minute episodes very interesting, if access is not restricted outside of UK:

    (you need to catch it quickly as the first episode gets taken down in a couple of days)