Monday, February 3, 2014

Phillips Curve Fallacy


There is a good article in Forbes about the Phillips curve fallacy that most economists seem to operate under.

Phillips studied wages under a gold standard and found that when labor was in tight supply that wages went up.  This is just saying that when supply is tight the price goes up.  This is basic economics and common sense.  

But most modern economists take this study and think that if they print money that employment will go up.   This study does not show that at all.

91 comments:

  1. Vincent, since that article specifically mentions NGDPLT and Scott Sumner, I think I'll post a link on themoneyillusion.com and see if Scott has a response.

    A few comments from my amateurish view:

    1. This statement from the article seems highly debatable:
    "The best most sensitive and reliable measure of the real value of the dollar is how many ounces of gold it will buy. This is because gold has maintained a reasonably constant real value over the centuries, and because gold is considered by many to be the ultimate and true “money.”"

    Really? Not being a gold user, I'm wondering how oz of gold varies against things I DO buy and use everyday like electricity, groceries and gas? Regarding emotional attachments to gold (the point of his last sentence?), I don't have one.

    Even if I grant him that on average, over the centuries, gold has maintained its value, that seems to be of limited usefulness. I'm more more interested in its volatility over much shorter time frames (like years, decades and centuries, no to mention single days).

    Implicit in the above quote and from the article in general is a rejection of the concept of a demand shock causing problems. Or that aggregate demand can cause problems. Sumner addresses this again in a recent article, essentially arguing that all demand shocks are monetary shocks and it should be the goal of monetary policy to eliminate those:

    http://www.themoneyillusion.com/?p=26098

    That particular Sumner article is interesting because it explicitly puts a little daylight between himself and Mark A. Sadowski on that issue.

    But regardless of Sumner/Sadowski differences on that, I think they'd both agree that the gold standard or something similar with a fixed exchange/convertibility rate would be terrible monetary policy in that it would accentuate demand shocks, and those in turn can cause recessions and depressions. Perhaps the only exception is if a country on the gold standard was primarily involved in only the gold industry. Here's how Nick put it:

    "If a country was little more than one big gold mine, a gold standard might be a quite sensible monetary policy.

    Hmmm. I think that if the US (or Canada) had targeted the price of wheat, the Great Depression would have been a lot less bad than it was."

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/separating-real-from-nominal-shocks.html?cid=6a00d83451688169e201a3fcb39ac8970b#comment-6a00d83451688169e201a3fcb39ac8970b

    Actually, the post itself touches on the gold standard as well as nominal vs real shocks (just as Sumner's article does).

    ReplyDelete
  2. OK Vincent, I put it to Scott here:

    http://www.themoneyillusion.com/?p=26098&cpage=1#comment-317171

    Also, one other comment on the article: he'd improve his chances of convincing people (rather than just speaking to those that already share his views) if he'd drop the dog whistle terminology. Specifically I'm referring to his repeated use of the term "central planner" which seems designed by political marketing consultants to invoke images of Soviet era communist "central planners" deciding on the price and quantity of every single good in the economy.

    I hate dog whistles, both right and left. Some of my least favorite: bankster, oligarch, rentier, statist, socialist, collectivist, Ponzi, etc. If used accurately, I have no problem, but they're typically used to exaggerate via their connection to other ideas. I react to them in the same way I do to the phrase "100% Free!!!" or "All Natural" or any other tired marketing slogan.

    ReplyDelete
  3. ... Also, he uses the term "Keynesian" in the same way that David Stockman does: as a catch-all for the bad guys.

    ReplyDelete
  4. Tom, what do you think of the main point. The Phillips curve was for wage rates under a gold standard, they went up when labor was in short supply. Duh. Now most economists use this to justify printing money. It is wrong to draw that conclusion. Right?

    ReplyDelete
  5. Sadowski replies (right under my question):

    "Tom Brown,
    I used to comment on Louis Woodhill when his articles were featured at Real Clear Economics. I’ve wasted far too many words on him already.

    Louis Woodhill:
    “This is because gold has maintained a reasonably constant real value over the centuries, and because gold is considered by many to be the ultimate and true “money.””

    Note that Woodhill puts scare quotations around the word “money” because some of the things that people call “money” aren’t real money backed by real gold."

    ReplyDelete
  6. Vincent, you write:

    "Tom, what do you think of the main point. The Phillips curve was for wage rates under a gold standard, they went up when labor was in short supply. Duh. Now most economists use this to justify printing money. It is wrong to draw that conclusion. Right?"

    I don't know. For me to answer fairly I need to re-read the article and maybe do a little homework. Hopefully Sumner will answer and not ignore the main point and spare me the effort. :D

    My main problem is that the author seems confused. He seems to deny that AD shocks caused by nominal monetary shocks can be a problem (otherwise he wouldn't be endorsing the gold standard). At the same time he's almost advocating NGDPLT as the best thing available outside the gold standard. I question if he even understands that AD = NGDP. Doing NGDPLT by whatever method will almost invariably result in some inflation as NGDP growth rate = inflation rate + GDP growth rate. Do you get the impression that the author understands this?

    ReplyDelete
    Replies
    1. Also I don't think I've ever seen an MMist use a Philips curve argument to justify inflation.

      Delete
  7. Vincent, more good stuff from both Sadowski and a little from Sumner on this starting here:

    http://www.themoneyillusion.com/?p=26098#comment-317227

    Sadowski makes three comments in all (including the one above the above link I linked to yesterday) and Sumner two.

    What do you think of Sadowski's reply? Sumner's? I think Sumner is indirectly addressing the Philip's Curve question in his brief reply to me, don' t you think?

    Sadowski's response is quite extensive.

    If you want to to challenge them I'd do it fast before the topic gets too stale: things move fast over there.

    ReplyDelete
  8. Vincent, I had one go to spam I think, but you can erase it. I'll reprint the text here from the meat of Sadowski's response (broken into two parts):

    ---------- Mark Sadowski: Start Part #1 -------------

    Tom Brown,
    Both Woodhill and Tamny abscribe to Austrian Business Cycle Theory. They are very similar in their personal outlooks on life, being largely self educated, anti-empirical, libertarian goldbugs. It’s the anti-empirical aspect that by far bothers me the most.

    What Woodhill says about gold maintaining a constant value is of course totally false, but his anti-empirical nature means he is immune to the exposure of data.

    However, for what it is worth here’s some things I dug up from my notes.

    First of all, it’s not at all clear that that gold has a deflationary bias (which would be a plus in Woodhill’s eyes).

    The nominal price of gold fell from $850 an ounce in January 1980 to $272 an ounce in July 2000. A gold standard might have committed the US to a more inflationary monetary policy over that time period.

    It’s not even clear that gold is even a good hedge against inflation over extremely long periods of time. For example gold was 0.89 English pounds an ounce in 1257. By 1999 it had risen to 172 UK pounds an ounce or an increase of just over 193 fold. But historical estimates of the Retail Price Index (RPI) by Gregory Clark have it rising from 0.144 in 1257 to 74.0 in 1999 or up by nearly 514 fold. The real value of gold went down by 62.4% over 742 years.

    The real problem with a gold standard is not the deflationary bias, it’s the price volatility.

    There or those who blame gold’ recent price variability on the demand for gold ETFs. But gold ETFs have only been in existence since March 2003. There’s no evidence of greater gold price variability since their introduction:

    https://research.stlouisfed.org/fred2/graph/?graph_id=118123&category_id=0

    In fact gold prices appear to have had far higher variability during 1973-1983.

    And annual data on internal gold demand from the UK during 1858-1914 (when the price of gold was fixed) shows extraordinary fluctuations:

    http://research.stlouisfed.org/fred2/graph/?graph_id=118124&category_id=0

    Other than the supposed emergency use of fiat currency by the Spanish in a siege in 1491 during the Conquest of Granada there’s no known record of fiat currency in the West before 1661. So currency consisted almost entirely of gold or silver coins.

    The historical Retail Price Index (RPI) by Gregory Clark is continuous from 1264 to present. The 396 inflation rates computed on the basis of that index from 1264 through 1660 show double digit inflation in 57 years and double digit deflation in 42 years. There are three years where prices rose by more than 30% and six years where prices declined by more than 20%. So there is little evidence of stable prices before the introduction of fiat currency.

    ReplyDelete
  9. -------------------- Mark Sadowski: Part 2 ----------------------


    Gregory Clark also estimates of average annual real earnings covering the same period of time. Real earnings fell in 197 years or nearly half fo the time. There are 59 years where real earnings fell by more than 10%, ten by more than 20% and one where real earnings fell by 30.8%. Through the entire 396 year period real earnings increased by 4.4% or at an average annual rate of increase of 0.011%. In short, real earnings were highly volatile but essentially stagnant prior to the introduction of fiat currency.

    Goldbugs often credit the rise of the US as a world economic power to the fact the US dollar was backed by gold and silver in the 1800s.

    The occupation of a lightly populated continent by a technologically superior civilation was bound to lead to some economic growth, no? But if I had to pick one reason for why the US grew from being a medium size power (it was already large enough to win its independence from one of the greatest powers on the earth) to being the world’s leading economy the last thing that would cross my mind is a silver and gold backed currency.

    The United States had five major deflations during the 19th century: 1801-1802, 1814-1821, 1822-1824, 1841-1843, and 1865-1878. Four of these are associated with “panics”: the Panic of 1819, the Panic of 1837, and the Panic of 1873. The average rate of real GDP per capita growth during these five deflations was 0.6%. The average rate of real GDP per capita growth during the the 19th century as a whole was 1.4%.

    Since the beginning of the 20th century the US has only had one major deflation: 1920-33. This was of course associated with the Great Depression. The average rate of real GDP per capita growth was (-1.0%) during those 13 years. Real GDP per capita growth has averaged 1.8% since the beginning of the 20th century (including the Great Depression) but has been even higher since we gave up the medieval practice of using specie as currency.

    All data comes from Measuring Worth.

    ReplyDelete
  10. Hey Vincent, take a look at this from Mark:

    "I’m looking at the real price of gold. And in any case, as I point out, the main problem with gold is the fact that its real price is enormously volatile, so the long run trend is totally immaterial." - Mark S.

    That was exactly my point to you (except that I was just pretending that I knew that... whereas Sadowski can probably dig up the references).

    ReplyDelete
  11. Vincent! ... I posted a link to Mark's comment on Forbes, and now both the author of the Forbes article (Louis Woodhill) and Sadowski have made additional responses at Sumner's site:

    http://www.themoneyillusion.com/?p=26098#comment-317317

    http://www.themoneyillusion.com/?p=26098#comment-317322

    http://www.themoneyillusion.com/?p=26098#comment-317364

    Stirring up trouble can be lots of fun! :D

    ReplyDelete
  12. You'll note that even Bill Woolsey gets in the act a bit too (on that same page).

    ReplyDelete
  13. Vincent, Sadowski has more in response to Louis now.

    Also, it sounds like Louis is proposing a gold standard but w/o direct convertibility. How is that different than the "Ponzi" gold standards that you've pointed out?

    ReplyDelete
  14. The gold standard that makes sense is gold coins where fractional reserve banking is illegal for fraud reasons. If a bank wants to lend for 10 years it needs to sell a 10 year bond so it has the money for that long. Lending for 10 years using demand deposits should be a crime. Then you would have stable prices under gold. Over thousands of years gold has been about the value of a nice suit and shoes per ounce. Fractional reserve banking causes inflation when it is booming and deflation when it is breaking, until you get a central bank as lender of last resort and you just get inflation all the time.

    ReplyDelete
    Replies
    1. In the 1950s a silver dime could buy you a gallon of gas. Same is true today. Not so with paper money.

      Delete
    2. Vincent, even if it were true that over 1000s of years the average value of gold has remained constant, that's totally unimportant to those of us that live 80 years or so. We don't care about long term (1000s of years) stability, we care about human lifetime and less time scale volatility.

      I'd be a lot happier with a dime that depreciated in value at a constant known rate forever. It's all about predictability: predictability is stability.

      Delete
    3. "The gold standard that makes sense is gold coins where fractional reserve banking is illegal for fraud reasons. If a bank wants to lend for 10 years it needs to sell a 10 year bond so it has the money for that long. Lending for 10 years using demand deposits should be a crime."

      That sounds like a statist/anti-capitalism approach. :D (sorry I couldn't resist!)

      But seriously, please explain how that removes the volatility of the value of gold? I don't see the connection at all.

      I don't want the value of my MOA based on wedding gift fashion trends in India.

      Delete
    4. "Vincent, even if it were true that over 1000s of years..."

      You've no doubt heard the joke about the statisticians that drowned to death in the lake that was on average only 6" deep...

      Delete
    5. The problem in the Ponzi Gold Standard was not the gold, it was the Ponzi. They were printing more paper than gold and getting a boom in the 1920s and then people started to get worried that there was not really enough gold in the bank for all the paper and as they took out the gold the law of 2.5 times as much paper as gold meant they had to reduce the money supply. The fractional reserve banking effectively lets banks make it seem like there is more money when they are making more loans and reducing the money supply if people are deleveraging. Also, the banks go bust. This makes for an unstable value of paper money. If you did not have fractional reserve banking and people used gold coins (not paper with 40% gold backing) then the value would be much more stable.

      Delete
    6. Fractional reserve banking is based on a lie. They tell all the depositors they can withdraw their money at any time and then loan out 90% of the money long term. If more than 10% of the people want their money they can not give it to them. So it is really a promise that they can not keep, or a lie.

      Delete
    7. "Fractional reserve banking is based on a lie. They tell all the depositors they can withdraw their money at any time and then loan out 90% of the money long term."

      Say everybody in the US announced tomorrow they would withdraw all their checkable deposits in one month. What do you think would happen? I'll admit it wouldn't be pretty, and the banks would not be happy, but I don't see a fundamental reason that couldn't happen: The Bureau of Engraving and Printing (BEP) would crank up production of paper money, the Fed would purchase it for its production cost, and then the banks would buy that paper money for face value from the Fed in anticipation of a huge demand for paper money. Where would they get the money to buy the cash with? They'd borrow it from the Fed. What would happen to the banks? They'd replace the liabilities on their balance sheets: instead of paying checkable deposit holders 0.01% a year, they'd now have loans of reserves from the Fed that they'd have to pay 0.25% a year on: They'd lose 0.24% of their spreads (the spreads on the interest rates between their assets and liabilities). That's about it. Also they'd try to convince as many people as possible to transfer checkable deposits to time deposits (savings accounts, CDs etc). The Fed would HAVE to loan them the reserves that they'd exchange for cash because if they didn't the banking system would crash, and that's what the Fed is supposed to prevent. If the ATMs run dry, responsibility lands squarely on the Fed for that. That will never ever happen.

      How could banks make up that lost 0.24% in spreads? I.e. say their average spread was 4% prior to the mass withdrawal, and now they're only getting 3.76%. Well they could charge more for loans, pay less on savings and CDs... there's a number of things they could do. They'd be in a bind for sure, since it would take some time to improve those balance sheets: maybe there'd be some layoffs, etc, but I doubt it would destroy the banking system.

      BTW, I distinguish between checkable and time deposits because reserve requirement rules only apply to checkable deposits, but I guess I didn't really need to: The Fed would still accommodate if the public decided to withdraw their time deposits as well. This has almost nothing whatsoever to do with reserve requirement rules. I could have easily set this story in Canada, Australia or the UK which have a 0% reserve requirement.

      You seem to keep going back to the loanable funds model and forget that banks create the money and the Fed accommodates them.

      The banks you conjure up don't sound like normal banks, they sound like shadow banks, which have to accept deposits PRIOR to loaning out funds. But that's not how the system works. Banks are members of the Fed.

      Also the banks currently have a lot of excess reserves (ER), but in terms of balance sheets that doesn't change the story much: instead of going from paying 0.01% to 0.25% (w/o ER=0) they'd go from earning 0.25% to earning 0% (ER > 0): They'd still lose about the same amount in their spreads (0.25%).

      Delete
    8. So since the Fed is guaranteed to accommodate the public's demand for cash, whatever it is, then how is this system a lie?

      Delete
    9. Oh, and BTW, the Fed has other options: it could go on a massive expansion of QE instead of lending reserves... the effect would be the same: Banks would end up losing about 0.25% of their spreads after the mass withdrawal was made.

      Delete
    10. Re: fractional reserve banking: there's no difference between a mass deposit withdrawal and implementing a 100% fractional reserve requirement: The end result is the banks lose 0.25% of their spread. However, the banks are perhaps slightly better situated to make some of that back with a 100% RR: they'd have a bit more leeway to entice people out of checkable and into time deposits (they could up the rate they paid on time deposits). Plus they could even attempt to charge depositors for the convenience of keeping a checkable deposit, and thus mitigate the loss of that 0.25% further. In summary, even a 100% RR would not restrict free market credit growth. In fact you could even have and RR > 100%. The more you raise it the more you make life difficult for the banks, but there's no fundamental threshold beyond which you eliminate or restrict credit growth: again, they're banks, not shadow banks! They are very different things. Shadow banks live with a fundamentally different kind of restriction. I really think you are confusing the two.

      Delete
    11. ... as Sadowski pointed out today to me, there's currently more than enough ER ($2.5T) to cover should all demand deposits ($1.5T) be withdrawn, thus the banks would essentially lose the IOR, but that's it.

      Delete
    12. Tom, the original Fractional reserve banking was a lie. The hack of having a central bank print money when the banks get in trouble means that after a banking crisis you eventually get a currency crisis. The cure has a horrible side effect.

      Delete
    13. But I don't think withdrawing all the deposits in cash would put the banks in peril. They wouldn't need to be bailed out just because all their deposits were withdrawn as cash. They might get in trouble for other reasons, but I doubt that would be one. We haven't had a serious bank run in the US since the gold standard days (and there were PLENTY back then... in fact they were a fixture of banking for centuries!). Cyprus is about the only example I can think of recently where a nation's banks were all in danger of runs, but then they effectively have a "gold standard": an ECB which is hyper concerned with inflation, and which Cyprus has almost zero influence over. Independence of the CB is one thing, but they've given their sovereignty away. Let that be a lesson to us. :D

      Delete
    14. I get your point though: you didn't mean modern "fractional reserve" banking (which really doesn't exist in the US, Canada, etc, since modern CB priorities make the money multiplier a myth and bank runs a thing of the past)... you meant the old convertible gold standard days fractional reserve banking: when the money multiplier actually meant something (and which I'll grant you could return at any time, the moment the CB decides to target a fixed quantity of base money rather than something more sensible).

      Delete
  15. Vincent, this has now expanded into a Cullen Roch post!

    http://pragcap.com/debunking-the-myth-that-a-gold-based-monetary-system-coincides-with-higher-growth

    One that I'm sure is going to be popular!

    ReplyDelete
    Replies
    1. ... and BTW, don't waste your time trying to convince me!... you can mix it up w/ Sadowski here... and Sumner, and now Cullen! Go for it!

      Delete
  16. Vincent, you might like this: Nick Rowe imagines a world with red money, which represents negative value. He says you can imagine it's "backed by garbage."

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/negative-money.html

    ReplyDelete
  17. Vincent you write:

    "The problem in the Ponzi Gold Standard was not the gold, it was the Ponzi. They were printing more paper than gold and getting a boom in the 1920s and then people started to get worried that there was not really enough gold in the bank for all the paper"

    But Mr. Woodhill seems to be suggesting that the Fed keep 0 oz of gold for backing: all they'd do is target $1 to a certain number of oz of gold, presumable using stand OMOs like they do now. People will suspect, ... er... rather KNOW for a fact that there's absolutely no gold backing their paper money, and that the Fed will thus not give them anything for their $1 except another $1.

    In our current system, gold is not the MOA and there's a free market for gold, and yet it's value in relation to other goods and services shows it to be volatile. You're suggesting that this volatility can be reduced by introducing a fixed oz convertibility and eliminating fractional reserve banking? I still don't see how those two steps would reduce the volatility of gold over what it is right now as it trades on the free market. I'd think gold bugs would be interested in keeping a free market for gold rather than allowing it to be intimately intertwined with the state's monetary system and other statist controls and draconian restrictions and regulations on free market credit creation.

    ReplyDelete
    Replies
    1. It is not a question of what we want. What we expect is that as more and more paper money is printed that more wealth will flow into the money that is not printed, gold and silver. I expect the paper reserves at banks will become worth far less and so their gold reserves will be seen as their only real reserves. Suddenly the central banks of the world are all on a "real bills" system with gold backing their currencies, with values for the currencies adjusting to their money supply divided by their gold reserves.

      Delete
    2. You make it sound like gold and silver are the only assets of value that a nation possesses. I don't think that's true at all.

      Delete
    3. I think ultimately a nation's CB is "backed" by the nation (in the US case, ultimately Tsy would have to bail it out should it get in trouble), and all that nation's assets: e.g. an ocean full of nuclear subs loaded w/ SLBMs, land, marine areas, air spaces, mineral rights, EM spectrum, infrastructure, the rest of the military, national labs, the power to tax, a productive tax base, etc.

      Delete
    4. The moment gold is made the MOA, it's value will no longer float freely on the market in relation to all other goods and services. So perhaps this is the best of all worlds right now!... you're free to take your fraudulent government script and exchange it for real money (gold and silver) immediately, and you can have confidence that the value of real money isn't being distorted due to being the MOA: use the funny gov paper money as a quick and temporary MOE only, and then store your wealth with real money.

      Do you set your rates in terms of real money? oz of gold or silver? There's no reason you couldn't.

      Delete
    5. "The moment gold is made the MOA, it's value will no longer float freely on the market in relation to all other goods and services." ... because of sticky wages and prices. Here's a recent bit on the former (worth reading): http://uneasymoney.com/2014/02/06/why-are-wages-sticky/

      Delete
    6. Tom, if all the paper money gets hyperinflation over the next few years, as I expect, the only central bank reserves that will matter are gold and silver.

      Delete
    7. So Vincent, what % of your personal savings do you put into gold and silver?

      I know we've discussed this before, but what's a "few" years? Does five work for you?

      Delete
    8. Not really willing to talk about personal stuff. If I get a bigger yacht some day I might mention it. :-)

      Few is a vague term because the truth is I don't know. I will be amazed if Japan lasts another year or two.

      Delete
    9. OK, Fed 2016 it is then: Japan in hyperinflation by then, right?

      Delete
    10. If Japan does not have hyperinflation by Feb 2016, I promise to be amazed.

      Delete
    11. Haha... OK, I'm holding you to it!

      Delete
  18. Vincent, if the Fed started trading it's Tsy bonds for gold and silver (and maybe platinum and other metals too?), $ for $, until all of its assets are precious metals, but it still has the same sized balance sheet, what in your view will be the consequences? Any?

    Is QE neutral as long as it's only accomplished by buying precious metals?

    ReplyDelete
    Replies
    1. I think the "Real Bills Doctrine" is correct that a central bank can issue notes if it gets real value for them that could later be traded back for the notes. So if all they did was issue notes for gold, from the beginning, they would have been alright. If they tried to move to that now it would probably cause the price of gold to shoot up and the value of the bonds they had to crash. If some country with a balanced budget started only issuing new notes when they bought gold, I expect their currency could become very popular and they could end up with a huge amount of gold. I am not sure there are any sizable countries with balanced budgets though.

      Delete
    2. balanced budget: there are some mid sized countries with those, right? Australia? How about in SE Asia? Singapore? (I guess that one's pretty small). I don't know either.

      Accumulating gold was the goal of mercantilism, wasn't it?

      Delete
    3. "If they tried to move to that now it would probably cause the price of gold to shoot up and the value of the bonds they had to crash."

      What if they just started doing this gradually, maybe a conversion to all precious metal assets w/in 50 years or so.

      50 sounds like more than "a few" and certainly is more than two (in the case of Japan), so I'm guessing you wouldn't think that would work.

      And regarding acceptable central bank purchases, what about land? That's as durable as gold, plus they don't have to worry about theft.

      Does land have a similar volatility in relation to other consumer goods as do precious metals? I don't know.

      Delete
  19. Vincent: O/T ... one of your concerns (if I understand you correctly) is that the Fed might have to raise interest rates to "slow velocity" say, but in order to do that they'd need to unwind their balance sheet first, correct? (Since ER > 0 implies that the FFR is driven down to the IOR). So in order to get FFR > IOR, they need to unwind.

    A couple of thoughts: first off, with ER > 0 they can always set the FFR with the IOR.

    Now if they really want the FFR > IOR instead, then a quick alternative to unwinding their BS (and flooding the market with bonds) is to simply eliminate the ER (set ER = 0) by raising the reserve requirement to 167% ($2.5T of reserves / $1.5T checkable deposits). They wouldn't have to sell a single bond to do that. What do you think?

    ReplyDelete
    Replies
    1. ...or did I get that wrong... maybe raising rates raises velocity. :(

      Well anyway, I like my idea of setting RR = 167%... perhaps it's a solution in search of a problem, but I kind of like it: it's allow a very quick way to set FFR > IOR should that be needed w/o doing any unwinding.

      Delete
    2. The problem is that as prices start to go up about all they can really do is raise the interest rates to try to reduce the rate they are making money. But this is not easy because as interest rates go up the velocity of money goes up. So in spite of their very painful efforts to fight inflation, inflation will continue. So it will look like the Fed has lost control of inflation and the money supply. And they will have lost control.

      Delete
    3. But recall that the Fed also can change the IOR. Nick Rowe describes the IOR as a direct way to control the demand for the MOA. So the Fed can increase the FFR above the IOR by manipulating the RR so that OMOs again allow arbitrary placement of the FFR above the IOR, but the IOR can be increased too, which directly controls the demand for MOA. Take a look at this and see what can be done when the Fed actively uses both controls (well all three really: FFR, IOR and RR):

      http://catalystofgrowth.com/theory/fiat-currency-construction/comment-page-1/

      Thus the quantity of base money can be targeted at the same time that NGDP is targeted: in other words, velocity can be targeted directly.

      Delete
    4. ... actually, to be fair, I think the rate Nick was thinking of was the rate in his special world, which is the universal deposit rate. I'm not sure if his world included cash or banks. Thus I think it was more akin the DOB's world. So IOR may not accurately describe the "demand for MOA" unless it applied to cash as well somehow.

      Now Nick differs sharply with DOB (DOB is an NKer) ... to Nick (as a monetarist) all that's really important macroeconomically, is the IOR rate (which he calls "Rm"). He explains it here:

      http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/two-interest-rates-and-a-simple-question.html

      Delete
    5. ... but still, I would think that IOR would have a definite effect on the velocity.

      Delete
    6. One last quick note on the powers of the CB: you always refer to the idea that bank reserves could leave the banks in the form of cash. But that's only true if the CB puts in the order for that cash to be printed/minted, and then agrees to sell it to the banks (and maybe even loans them the money to buy it with --- at least pre-2008 -- currently the banks have more than enough reserves and wouldn't need any CB loans). But the CB could shut that down at any time by refusing to do that (i.e. they could let the ATM machines run dry). Would it happen? I doubt it, but I suppose in an emergency it's still an option.

      Delete
  20. Vincent, you might like this: I asked Nick Rowe to summarize what the goal of good monetary policy should be. His response:

    "...the goal of good monetary policy is to try to make Say's Law true."

    ReplyDelete
    Replies
    1. If you outlaw fractional reserve banking (bank must sell 10 year bonds to get money for 10 year loans) and use gold and silver coins, Say's law would be true.

      Delete
    2. If you accept the concept that shocks can occur to the demand for the MOA (or that there are nominal shocks or aggregate demand shocks), and that prices and wages are sticky, then I don't think that's true.

      Say you had a barter economy and gold was a minor part of the overall market and 1 oz = 1 chicken = 1 duck = 1 hr of labor. Now there's a shock to the demand for gold (demand doubles overnight) and suddenly 0.5 oz = 1 chicken = 1 duck. No biggie: the gold market has been affected, but relative values in the rest of the economy have been preserved (i.e. 1 duck still equals 1 chicken still equal 1 hr labor, etc.).

      Now go to a monetary economy, and make gold the MOA with $1 = 1 oz. If the same shock occurs, then by definition $1 = 1 oz still because gold is the MOA. Instead all the other prices & wages in the economy must adjust. If prices & wages were not sticky, then this happens instantaneously and there is no problem. If 1 hr of time used to be $1 prior to the shock, then 1 hr of time after the shock is now $0.50. But if wages & prices are sticky, then this doesn't happen instantaneously: it takes time, but eventually 1 hr = $0.50. And 1 duck = $0.50 = 1 chicken too. But in the near term, destructive transients are introduced into the economy that take time to work through: employees don't accept that they will only get $0.50 for 1 hr of time, and chicken and duck farmers don't accept that they will only get $0.50 cents for their fowls. So instead the firms lay off half their employees because there's no escaping it: gold has doubled in value, and they can't afford them anymore. Interest rates change too and production falls. There are destructive transients introduced because money is weird and prices & wages are sticky.

      What would a smart central bank due in such a circumstance? They would immediately devalue the money: essentially introducing an offsetting shock: they'd revalue $1 = 0.5 oz. What's the advantage of this? Wages don't change, prices don't change... all that changes is one price in the gold market, much like in the barter economy case: the value of the MOA is instantaneously changed in relation to everything else and no destructive transients are introduced.

      That's my interpretation of the MM story of recessions and how a CB can take action to preserve Say's Law and avoid unnecessary destructive transients in the economy due solely to nominal shocks combined with wage & price stickiness.

      Did you like my story? :D

      Delete
    3. Just finished refreshing my limited knowledge of Say's Law (yes, wikipedia). From what I gather it boils down to "general gluts cannot exist" ... and this is true (according to Say) because it's not rational for money to be hoarded (or for there to be an excess demand for money). I don't buy that at all!... BTW, Rowe says he got his quote from Brad DeLong... and it's true Brad often says something like that: Monetary policy should try to make Say's Law true in practice (but of course it's not true in theory). The MM story about an excess demand for gold (in a gold standard system) leading to a "general glut" I think does demonstrate a departure from Say's Law in practice. And if the CB is powerless to revalue the UOA (say it's a fixed gold standard), then these general gluts cannot be eliminated, and Say's Law "in practice" remains unattained. It seems to me that a fixed conversion rate gold standard could in fact be unusually susceptible to this kind of "general glut."

      Delete
    4. Vincent, as you know I'm an amateur hack... but Nick did bless my story above and Scott had this to say:

      "Tom, Your friend [that's you Vincent!] is wrong, as you’d still have NGDP shocks and sticky wages."

      So basically the same story I told above (only with a lot less words). So I don't know if the MMs are right, but I think I captured their objection fairly. And it's not just the MMs... after reading a bit from Brad DeLong's posts, I think he'd share that view, and perhaps Keynes too (Nick points out). The wiki article on Say's Law gives the breakdown of what different schools think about it. It sound's like you're in the general camp that doesn't believe that it's true in theory, but can be made true in practice. Nick tells me even Say himself was in that camp later in life. Of course there's a lot of variation w/in that camp as you, Nick, Scott, and Brad DeLong are all members.

      The other camp, of course, says that Say's Law is true in theory and practice all the time, which I can't imagine being correct except maybe in a genuine barter economy.

      Delete
    5. I believe that fractional reserve banking and central bank money creation are the cause of the booms and busts that you see. If it were not for these two things, then money would be stable and you would not have to worry about sticky wages or NGDP shocks. It would sort of be like a genuine barter economy.

      Delete
    6. Even w/o any banking at all, just because there is money used instead of barter opens you up to the possibility that there could be excess demand for money itself, which combined with sticky wages & prices steers you well away from the barter ideal. Say (at first anyway) didn't think excess demand for money was possible. Money is weird: if you have N goods in your economy (including money), then you have N-1 money markets. It's value literally affects every other market. Nothing like that exists with barter: the value of each good, only affects the value for that one good: a shock to the supply or demand for any one good does not disrupt the whole economy. Not true with a money economy: there is exactly one good whose value has the power to disrupt the whole economy: money itself.

      Delete
    7. BTW, booms and busts are different than bubbles and popping bubbles. Because Sumner believes in the EMH he claims bubbles don't exist, but he does acknowledge that booms and busts exist. Even if Say's Law is achieved in practice, this will not protect the economy from booms and busts (according to Sumner). What's the distinction between a boom and a bubble?... I can't tell you off the top of my head, but a boom is an economy wide phenomena I think, whereas a bubble is usually in reference to one market segment: "the housing bubble" for example. I asked Sumner once... but I don't recall his response precisely. It makes sense to me: I don't see why a purely barter economy would be free of ups and downs.

      As for eliminating fractional reserve banking and central bank money ... I don't see why that would eliminate booms and busts any more so than going to pure barter would.

      To see that money based on gold would not be stable you need only look at its volatility in relation to other goods and services as it stands now: as a pure commodity, like every other. That volatility would translate directly into MOA demand shocks if it were made the MOA, again, even w/o banking.

      Then the only other question is if wages and prices are sticky. I think it's clear that they are.

      Let's take banking off the table for a moment: say it's banned altogether. You'd like to base the value of our MOA on a single commodity. You think that'd be more stable. I'd like to see the value of the MOA divorced from any one commodity and instead have it's value determined in a way specifically designed to appear as if money was not a commodity (i.e. to make it appear that we have a barter economy). I think that would combine the best of both worlds: eliminate problems associated with excess demand for the MOA (the problem of money) and also eliminate the need for a coincidence of wants (the problem barter has).

      Delete
    8. BTW, my story about the ducks and chickens didn't depend on there being fractional reserve banking, nor did it depend on their being a central bank: it depended solely on there being a clearly defined UOA, and MOA, and the possibility of volatility in the demand for (value of) the MOA. And sticky wages & prices of course. If you're asking us to gamble on the non-volatility of gold, I think that's a bad bet.

      Delete
  21. Here's Scott on booms vs bubbles:

    "Tom, A boom is when output exceeds the natural rate. Bubbles don’t exist, but if they did it would be when prices were clearly above true value, and anyone could get rich simply by selling short and waiting a decade or so."

    http://www.themoneyillusion.com/?p=26038#comment-315595

    ReplyDelete
  22. Vincent, I apologize for leaving this many comments. Shees!... I was going to erase some and try to consolidate, but it's not worth it.

    But what I hear you saying is that you think that if FRB was outlawed and gold was made our fixed convertible MOA (i.e. fixing a UOA at some oz / $ and forcing the CB to buy gold as the only means for issuing script), then gold volatility would disappear, or become insignificant. Is that right? Would the whole world have to do the same, or do you think this could work if just one country did it?

    ReplyDelete
    Replies
    1. i.e., I don't hear you disputing wage & price stickiness: just the volatility of gold, should your advice be followed. And should your advice be followed, then it would take both stickiness & gold volatility to undo Say's Law (in practice).

      Delete
  23. Vincent, I asked Sadowski about what the evidence has to say regarding unwinding QE, specifically in regards to yields (he'd previously left a comment about the effect of QE itself on yields). His responses start here:

    http://www.themoneyillusion.com/?p=26159&cpage=1#comment-318434

    Here was his previous comment on QE and yields:

    http://www.themoneyillusion.com/?p=26159&cpage=1#comment-318301

    Within the response he leaves a link to a previous Sumner article based on other comments of Mark's regarding Sweden vs Denmark on the same subject. I thought it made for some fascinating reading. His conclusion: don't unwind QE. Do you take issue with his evidence (granted there's not a lot of it)?

    ReplyDelete
    Replies
    1. So only Japan has really tried to withdraw QE and it seems clear this contributed to their downturn.

      I like him saying that QE increases long term inflation expectations. "... why anyone would expect QE to “significantly reduce long-term yields” is beyond me.". Printing money and buying bonds can reduce interest rates in the short term, but long term it increases them. It is a short term fix with a long term problem. Mark is a smart guy. :-)

      Delete
    2. Oh, the point being that the odds are the US will never withdraw their QE. So eventually it will cause inflation. Only question is how long till "eventually" gets here.

      Delete
    3. Vincent, you should also read this:

      http://www.themoneyillusion.com/?p=26159&cpage=1#comment-318483

      I go on to ask him if raising reserve requirements (RRs) is a good alternative for putting the brakes on inflation since he says that raising the RR is contractionary. Offhand it seems like it would be since it involves no OMOs (no BS unwinding) and no payout of IOR.

      Plus if we raise the RR to 167%, nobody could claim we had a fractional reserve banking system, at least not one with a fraction less than 1. :D

      Also, you write:

      "Printing money and buying bonds can reduce interest rates in the short term"

      but did you see the chart he included?:

      http://1.bp.blogspot.com/-BAnvd5XkIkc/UnKFj667RyI/AAAAAAAAAyM/-xbQBsjqcm0/s1600/QE+Rates.png

      I don't see even a short term affect there. Perhaps smart people at the Fed realized that was never the goal of QE: it was always about long term expectations, which resulted in raising rates. What do you think?

      Delete
    4. Vincent, it seems like you've argued before that the Fed might be forced to unwind their BS, which leads to a hyperinflationary feedback loop. But now you say that if they don't ever unwind their BS, then this will also lead to inflation (hyperinflation or just inflation?). Some people would be happy if it led to a little more inflation (as that might help get NGDP back on trend).

      Also notice that Mark documents cases where GDP declined when QE was unwound... and those cases didn't involve hyperinflation.

      Delete
    5. Personally I think that between declining deficits, and the Fed's control over IOR, OMOs and RRs (and capital requirements too), things look good for avoiding out of control inflation. I guess we'll see. Plus there's a new "reverse repo rate" tool:

      http://www.piie.com/publications/interstitial.cfm?ResearchID=2558

      which I don't understand yet.

      Delete
    6. Tom, rates went from 4% down to 1.5% on 10 year bonds, lowest rates in forever, even back when they were on a gold standard, and you don't see any effect? You think this 1.5% is a natural free market interest rate for a currency that is being printed like crazy?

      The right thing to do, as Peter Schiff often says, is to stop printing money an accept the recession that will follow. That is not what I expect the US to do. I expect they will keep printing money.

      The feedback loops involve more money printing as prices for government obligations go up (poor people, retired people, government employees, etc) not less money printing. It is positive feedback loops that blow up. Negative feedback loops are under control.

      Delete
    7. I think Sumner would say that low interest rates are a sure sign of tight money. Notice in that graph the interest rate dropped every time a QE program ended.

      I also don't think Peter Schiff has our best collective interests in mind. He just seems like a constant drum beat of negativity. He gives me the same fuzzy feeling inside as those emails I get from Nigerian princes, needing my bank account to transfer multiple million dollars with. Sometimes he'll be right (the broken clock scenario) and when that happens he makes as much hay out of it as possible. The rest of the time, he comes up with excuses like inflation doesn't have anything to do with the price level, it's only about the base money supply, and therefore he's been right about inflation all along.

      I don't buy the argument that what we really need is a good recession, because it'll be good for us. Good for him maybe, but not the rest of us. Is that really what Schiff thinks?

      Delete
    8. QE has not ended.

      What Schiff and I both think is that if the government switched to living just on the taxes they collected there would be a recession but they would avoid hyperinflation. If they keep going the way they are going (which we expect) they will eventually get hyperinflation. We think in the long run you are better off choosing the recession but that since on any given week that looks worse, they will keep trying to kick the can down the road by printing more money till we get hyperinflation.

      Delete
    9. suffering through a recession seem to me a destructive way for the economy to reach steady state again after an AD shock.... wages are sticky, but not immovable... eventually they and prices will fall, and we can get back on track again: after a lot of pain. That seems totally unnecessary and a huge departure from the Say's Law ideal.

      Delete
    10. "QE has not ended" ... the purchasing programs associated with each phase of QE. Look where QE1 and QE2 end:

      http://1.bp.blogspot.com/-BAnvd5XkIkc/UnKFj667RyI/AAAAAAAAAyM/-xbQBsjqcm0/s1600/QE+Rates.png

      Delete
    11. What you have to understand is that the shock to the system was all the new money created by both banks and central banks during the boom. Trying to keep this shock from having a hangover take more and more shock, till the currency dies. Best to just take the hangover. The more drugs you take the worse the hangover will be eventually.

      Delete
    12. I think all the big investors know it was never really going to end. In the long run, once they start printing like crazy they never really stop.

      Read the paper off to the right about the French hyperinflation. I think they made a QE1 and it felt good, then they needed more and made a QE2, then that wore off and they needed a QE-infinity. Then things fell apart. It is very similar really. It is a very good read.

      Delete
    13. ... maybe they will "kick the can down the road" indefinitely and never get hyperinflation. In fact I suspect that is what will happen... because "kicking the can down the road" just amounts to growing the monetary base, which will always happen if NGDP levels grow 5% a year (for example).

      I think we live in a dynamic world... and we should shoot for stability in trends and rates... not absolute values. I'm completely comfortable with an ever swelling monetary base as long as we avoid other problems.

      Delete
    14. I think Austrians would like us to have a disciplined economic system, and that's what reintroducing the gold standard is in some sense intended to do. Same w/ outlawing FRB.

      MMs are equally interested in discipline, but their form of discipline I think is an inherently better one: keep NGDP levels on trend, growing at 5% a year. It's not tied to the volatile relative value of some arbitrary commodity: a commodity very few of us actually use for anything.

      Delete
    15. In other words MMs like discipline. Austrians are more masochistic and prefer bondage and discipline. :D

      Delete
    16. Tom, Japan is doubling the base money supply in 2 years. The average 10 day period it goes up about 1.6%. They are so far from a disciplined 5% per year it is like a whole other planet. The US is similar really.

      Delete
    17. But wait: 5% a year in the growth of NGDP is not the same as 5% a year growth in the monetary base. Actually I'm partly with you regarding growth of the base... I suspect that there might be a more efficient way to achieve 5% annual growth in NGDP. I know that many MMists would agree with me too: that if the US or Japan would just explicitly adopt NGDPLT, for example, then they wouldn't have to do so much QE. But I'll even step outside outside of MM orthodoxy and state that perhaps ... just perhaps ... a smarter use of fiscal policy combined with monetary ... combined with being very explicit about the target, is really what's required. Shoot, I'm not even that far outside of orthodox MMism in saying that... David Glasner has suggesting something similar (regarding using fiscal too) and David Beckworth has on numerous occasions described his version of a "helicopter drop" (to be used only in extreme emergencies... Lol) which also folds in a fiscal element.

      Also, apparently Japan is one of the few oddball nations (along with the US) that has both an RR > 0, and IOR facilities: so they could conceivably pull a stunt like I already referred to, and raise RR > 100% to put the kabash on any run-away inflation. (Sadowski says that raising the RR is contractionary). That's very painless for the CB to do (although to be fair, it's clear to Sadowski favors using IOR instead of RR, but he acknowledges they can both do essentially the same thing)..

      Delete
  24. Vincent, I like this article by Sumner (it ties into our discussion of bubbles above):

    http://www.themoneyillusion.com/?p=26140

    ... especially the last bit about current account deficits.

    ReplyDelete
  25. Vincent, I'm simultaneously having an ongoing debate w/ Mike Sax (author of the blog "Diary of a Republican Hater") about whether or not being an MMist necessarily means you have to hate gov, want to lower taxes on the rich while imposing a consumption tax on the poor, etc. Mike seems to think that MM = you're really an arch neo-liberal conservative at heart, and want to "drown government in a bathtub and dance on Keynes' grave."

    Well he's had plenty of pushback on that from me, and now some from Nick Rowe and Mark Sadowski too (even a bit from Sumner once). But government hater Morgan Warstler rides to Mike's defense (as his unnatural ally) here:

    http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/tiff-macklem-retail-competition-flexible-it-vs-ngdplt.html?cid=6a00d83451688169e201a5116c2467970c#comment-6a00d83451688169e201a5116c2467970c

    He really likes to highlight the discipline aspect I was getting at earlier. :D

    ReplyDelete
    Replies
    1. Ok, I joined in.

      http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/02/tiff-macklem-retail-competition-flexible-it-vs-ngdplt.html?cid=6a00d83451688169e201a73d779bc0970d#comment-6a00d83451688169e201a73d779bc0970d

      Delete
  26. Tom, I answered you over on pragcap but am putting my answer here as well.

    If when the USA was founded they required banks to sell 10 year bonds to get cash to use for 10 year loans (can't use demand deposits, must have matched duration bonds, so banks could not make new money) and never went off a gold/silver coin money (no paper money or central bank) then I think that there would not have been anywhere near the number of inflations/booms/busts/deflations/crisis between then and now. If regular banks and central banks can make and destroy money then the money supply is not stable and you get crisis after crisis.

    ReplyDelete
    Replies
    1. I replied here:

      http://pragcap.com/about-that-1929-chart-the-details-matter/comment-page-1#comment-167562

      ... but I'll add that it's not the money supply itself that should remain stable, it's the value of the MOA that should at least move in a predictable way: the money supply should remain flexible. Because even a moderate volatility in the value of the MOA combined with sticky wages & prices moves you far away from the Say's Law barter ideal, and this is regardless of the stability of the stock of MOE. We are much better off having an MOA whose value is completely predictable, and a flexible stock of MOE. Any surprises in the value of the MOA are really the same thing as nominal shocks. Uncompensated nominal shocks mean you can forget about Say's Law even appearing to be true. Nick Rowe says it well here:

      "What is the problem with fluctuations in AD around that trend? Fluctuations in real activity, (plus maybe fluctuations in inflation)."

      Delete
  27. I made a new post for this thread to move onto:

    http://howfiatdies.blogspot.com/2014/02/preventing-crises.html

    ReplyDelete

Looking for polite debate on ideas. Never attack a person. Be nice.