Tuesday, January 14, 2014
Dangerous to back your currency with the future value of your currency
One of my many explanations of hyperinflation is the Backing View or Real Bills Doctrine. This says that the central bank has to have assets that it can use to withdraw the notes it issues to support the value of those notes. It also warns any bonds should be short term, like 60 days or less.
The Fed owns 40% of all treasuries over 5 years in maturity. About $3 trillion of the $4 trillion in assets the Fed owns are more than 5 years. This means that the Fed is using long term bonds, against the advice of the Real Bills Doctrine. The reason this is a problem is that it ends up backing the current value of the notes with the future value of the notes. If interest rates are going up the value of the bonds will go down and so the backing of the current notes becomes worth less and less. But the more the value of the current notes drops, the less the long term bonds will be worth also. This can spiral out of control.
The Fed is backing the dollar with the future value of the dollar. If this starts to go bad it can go really bad.
Subscribe to:
Post Comments (Atom)
"If this starts to go bad it can go really bad. "
ReplyDeleteVincent, on a scale of 0 to 10, how disappointed will you be if things don't really go bad? :D
Tom, it is like watching a train that is about to crash. It is not that you want it to crash but it is so interesting you can not look away.
DeleteAlso, there is nothing you can do about a train crash, except maybe try to get out of the way as much as possible. Same with this.
DeleteBTW, You can add a couple of more folks to argue with on their blogs (in case you haven't already): one's a traditional monetarist (John Cochrane), and the other a self proclaimed "New Monetarist" (Stephen Willliamson). Here's Cochrane:
ReplyDelete"In my opinion, QE has essentially no effect. ... But the good news is that we therefore can't worry too much about its reversal. It’s neither going to cause hyperinflation, nor need it cause much trouble when the Fed "tapers.""
https://www.richmondfed.org/publications/research/econ_focus/2013/q3/full_interview.cfm
and here's Stephen Williamson:
http://newmonetarism.blogspot.ca/2013/11/liquidity-premia-and-monetary-policy.html
...of course he caught a lot of flack for that one from across the spectrum:
http://noahpinionblog.blogspot.com/2013/12/does-qe-cause-deflation.html
So it seems the Monetarists in general don't think QE will cause hyperinflation, but instead either deflation (NM), nothing (TM), or be helpful (MM). :D
... and BTW, I've been posting links to your blog here Vince in several places, as an example of a non-insane hyperinflationist. Any new traffic sources for you?
DeleteA "non-insane hyperinflationist". :-) I am not noticing new traffic. You might have to work out a better sales pitch. :-)
DeleteThanks for the links. I commented on:
Deletejohnhcochrane.blogspot.com/2013/12/what-if-we-got-sign-wrong-on-monetary.html
with a link to my
http://howfiatdies.blogspot.com/2014/01/how-we-know-inflation-is-coming.html
But awaiting moderation.
Williamson was asking what has come from econoblogs and I commented here:
Deletehttp://newmonetarism.blogspot.com/2013/12/whats-economics-blogosphere-good-for.html?showComment=1390053617617#c8364510806425013788
Ya Tom, I really think you have to work on your sales pitch for my stuff. When I comment I am much more assertive about the correctness of my position than your "non-insane hyperinflationist" sort of sales pitch, and I get some traffic from my comments. :-)
DeleteWell sometimes I also mention that your word is good, and that you will pay if someone meets your $100 challenges!
DeleteYou did get people to come for the first round of that contest. Thanks! Does not look like anyone is taking the second round though, eh?
DeleteMaybe if you upped the prize? ;)
DeleteRegarding your post, Nick Rowe, Mike Sproul and I have had some discussions about this same thing in the past. See comments in:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/12/banks-with-100-capital-ratios.html
There's potential for a death spiral, as your post describes.
Thanks! In 4 years of blogging this is the first time someone else with an econblog has agreed with me! This is a historic occasion. I love the quote, "Unfortunately, central banks hold bonds denominated in their own shares (i.e., their own money), which is like holding calls on your own shares, and acts to destabilize their money through an inflationary feedback effect." Nice way to put it!
DeleteJP, just to make sure you are fully agreeing with me. Do you agree that people living through such a central bank death spiral would usually call it "hyperinflation"?
DeleteYes, I think so.
DeleteVincent, what do you think of the following quote and article:
ReplyDeletehttp://online.wsj.com/news/articles/SB10001424127887323353204578127374039087636
“The US most certainly is insolvent, as its unfunded liabilities amount to tens of trillions of dollars and are increasing at an astronomical rate. There is no possibility of these liabilities ever being paid off without immense inflation.”
I believe that is correct.
DeleteYour in good company with Greenspan and Stieglitz "The US can pay any debt because it always print money to do so" . Just a polite way of saying, they'll inflate their way out.
DeleteOur Aussie ex PM on from 2009 said something life this
TONY JONES, PRESENTER: You talk about a new Bretton Woods agreement?
PAUL KEATING, FORMER PRIME MINISTER: A totally new Bretton Woods agreement. We're not going to get out of this. I mean this is the United States Budget cannot reflate the world.
We've always lived in a position where the United States Budget could reflate the world this is not going to happen now. The Budget this year was going to be $850-billion, now look President Obama is talking about another trillion, so $1.8-trillion, their GDP is 13-trillion, so they'll be running a Budget deficit this year of 15 per cent of GDP, they'll have to do this for three or four years.
Sixty per cent of American GDP, who is going to buy the bonds? Now every serious American policy maker knows that they are not going to be returning value, in the end they'll inflate their way out. So in other words, you'll buy American Treasury bond, but what you get back in return will be an inflated dollar, so you'll get back 50 per cent of real value, or something, in other words the debt will be so overwhelming that it cannot be repaid.
And you'll start to see in the price of gold, if this goes on for a couple more years, the real serious question of an American default, a default by the United States Treasury. So this is what we are dealing with now.
The full vid is on youtube for anyone interested
http://www.youtube.com/watch?v=nkux40HUTtc
Vincent, here's what Sadowski had to say about the above Cox & Archer WSJ article and the quote I provided (about the article):
ReplyDelete"Assuming the $86 trillion figure is correct, it is the present value of the additional resources that would be necessary to meet projected expenditures at current law levels for all those programs combined for the next 75 years.
According to the Medicare Trustees report the present value of GDP for the next 75 years is $944 trillion (page 228):
http://downloads.cms.gov/files/TR2013.pdf
In short, yes it’s big, but the present value of GDP for the next 75 years is much bigger. Also, a lot can happen in 75 years. A small shift downward in cost projections and a small shift upward in revenue could make that scary sounding figure vanish completely."
Here's another take on it:
http://www.fool.com/investing/general/2012/11/29/how-to-scare-people-with-really-large-numbers.aspx
and some others, especially in the comments here:
http://blogs.law.harvard.edu/philg/2012/11/27/wall-street-journal-article-on-calculating-federal-government-liabilities/#comment-194118
All in all I think I have to agree with the critics of the WSJ piece. I wish they'd provided links to the documents they'd used and explained the math they did a little better. It seems they were perhaps willing the use gov documents like TR2013.pdf to come up with terrible sounding numbers for the so called "liabilities" (which, as the "Fool" article points out, are not real liabilities) while at the same time ignoring the big numbers for the GDP, that I assume were calculated in the same way and present in the same gov docs they used. Why do you suppose that is? I went through their figures looking at TR2011 and TR2012 as well and I didn't find a match on how they came up with the $86.8T or even any components of it. If you Google that WSJ article you'll find copies all over the place, but very few comments on exactly where the numbers came from and exactly why and how they do the calculations they do which seem, in retrospect, tailor made to scare people.
What are your thoughts? After reading Sadowski's response, does it seem fair that they talk about a "liability" 550% of GDP, while if one were doing a calculation of GDP that is perhaps more directly comparable to their "$86.8T" figure, makes their number less than 10% of "GDP."
I'd really like to know your thoughts... I'm having a hard time because I can't replicate the WSJ numbers, but the evidence seems to point to that article being purposely misleading. (BTW, I don't see why the "Fool" response article says that their numbers came from looking out to infinity either: I wish that author had explained WHY he thinks that. It's hard to believe it's really to infinity).
If the WSJ article was written to be purposely misleading, then it's a shame so many people just copied it on their sites w/o any critical review.
BTW, here's a couple of quotes from pg 227-228 of the TR2013.pdf (their pg #s, not the number you type in acrobat):
Delete"For comparison, the present values of HI taxable payroll, OASDI taxable payroll, and GDP are $428.2 trillion, $372.2 trillion, and $944.2 trillion, respectively, over the next 75 years. This present value of GDP is calculated using HI-specific interest discount factors and differs slightly from the corresponding amount shown in the OASDI Trustees Report."
The above is "Note 1." to "Table V.F2.—Present Values of Projected Revenue and Cost Components of 75-Year Open-Group Obligations for HI, SMI, and OASDI"
And then on the next page there's this:
"From the 75-year budget perspective, the present value of the additional resources that would be necessary to meet projected expenditures, at current-law levels for the three programs combined, is $39.6 trillion.106 To put this very large figure in perspective, it would represent 4.2 percent of the present value of projected GDP over the same period ($944 trillion)."
Tom, social security was supposed to be a pension fund where people paid into it and the value grew fast enough to pay people their retirement when they were old. But the rest of the government has effectively "borrowed" this money at near zero percent interest and spent it. So the obligations are going up much faster than the size of the fund. It is as if a company had a pension fund that was way under funded. A company would have to use GAAP accounting and show this increasing liability as part of their loss for the year. The government does not use GAAP. So really their deficit is bigger than they show. Yes, things could change and maybe it will all work out, but I think a betting man should bet on them printing lots of money to get things to "work out".
Deletehttp://en.wikipedia.org/wiki/National_debt_of_the_United_States#Unfunded_obligations_excluded
http://www.wnd.com/2011/03/278017/
I put "borrowed" in quotes because if I took your money by force and paid you near zero percent interest and you had no say in the matter, you might call it "stealing".
DeleteRegarding the rest of the government "borrowing" the money, you mean that SS, Medicare and the rest use undistributed funds collected each year to buy Tsy bonds. I don't see anything wrong with that. Should they just let the funds sit unused and uninvested in a Fed deposit account? I don't even know if the Fed pays 0.25% on those deposits (after all, it's called "interest on reserves" and Fed deposits not owned by banks are, by definition, NOT reserves).
DeleteIf intra-gov agencies (SS, etc) buy T-bonds, and the Tsy deficit spends or pays down principal on Tsy debt, then at least those funds are returned to the private sector instead of sitting idle. So I suspect that even if the Tsy were accumulating a surplus each year, they'd end up selling Tsy bonds to SS and using the proceeds to pay off maturing T-bonds.
Re: liability status: do you think that Motley Fool article is correct? It states that to call these unfunded "liabilities" is incorrect, since congress can change the law at any time, like Reagan and the congress did back in 1980s. It's very different that a local government pension fund. Also the Fool article had some interesting things to say about the Medical entitlements in this regard and how they never actually go into debt: services would just decrease. Take a look at the TR2013.pdf document, I think it discusses this some.
There's this, for example:
"Other than asset redemptions and interest payments, no provision exists under current law to address the projected HI and OASDI financial imbalances. Once assets are depleted, expenditures cannot be made except to the extent covered by ongoing tax receipts."
And I think that's why Sadowski mentioned projecting current laws in his response. These are not liabilities like a company or local gov would have, pension wise. Perhaps that's why GAAP doesn't apply?
Still, the WSJ article seems like scare mongering to me. They talk about these "unfunded liabilities" being 550% of current GDP, when they should actually state that they represent 9.2% (=86.8*/944) of the present value of projected GDP over the next 75 years to make the comparison fair. I don't think they were interested in presenting a fair picture though. Plus I still question that $86.8T figure. If you go look at the TR2011.pdf I think you'll find (for example) a figure of $33T rather than the $40T they quote in the WSJ article for one of the components (Medicare I think it was). I'd just like to see a complete piece with references to justify their math, but I suspect that would tend to undo the purpose the authors had in mind when writing the piece.
I'll grant them though that the $86.8T may be the correct order of magnitude for the present value of total projected shortfalls over the next 75 years, assuming no laws are changed, and of course that circumstances don't change.
Also, that WSJ article talks about the "cupboard being bare" on the SS trust fund, but I think that's just another scare tactic... as I discussed above, of course those funds will be used to purchase T-bonds, and the proceeds used for something, even when the Tsy is accumulating a surplus (i.e. paying down total debt).
It is not a liability in the normal sense as a change in law can change how much it is. But there is a real obligation to pay these people who have paid into the fund all their working life enough that they can live on. So there is only so much wiggle room really. If people on social security start starving to death there will be enough shooting to call it a revolution.
DeleteLook at this link to see what I mean about GAAP accounting showing more real deficit each year than the reported federal deficit.
http://www.wnd.com/2011/03/278017/
Vincent, re: "borrowing" from the SS trust fund, I'd rather they bought Tsy bonds than just let the money sit in a Fed deposit.
DeleteOf course you probably don't like that SS exists anyway, but assuming we take that issue off the table, and SS exists whether you like it or not, what would you do with the undistributed funds?
It should have been invested like a normal pension fund. Some money in bonds, some in stocks, some in real estate, some in gold. Investing at 0.25% just means they are getting further and further behind each year.
DeleteVincent, I'm scared of that wnd.com URL: that's "World Net Daily" isn't it? The guy that runs that (w/ the mustache) is a extremist lunatic isn't he? I have a security clearance I need to maintain... that's not going to shoot me off to some sort of pro-militia / sovereign-citizen / antisemitic / Aryan Nation website is it? I want to keep my NSA profile unblemished. :D
DeleteIf you assume me it's safe, I'll take a look.
"assume" should be "assure"
DeleteI assure you that your NSA profile won't get any worse after reading that page than after conversing with me.
DeleteMaybe I've already blown by NSA profile by visiting howfiatdies.com and referring to the author as "non-insane" :D
DeleteOK, last thing before I head over to wnd, to summarize my long winded comments above, and assuming the $86.8T is correct, what do you think is the proper % to talk about?
Delete1. $86.8T is 550% of the current GDP
2. $86.8T is 9% of the present value of total projected GDP over the next 75 years
Isn't the $86.8T *probably* the present value of the total projected shortfalls in those programs over the next 75 years? So why compare it to a totally different figure?
550% vs 9% is a pretty dramatic difference isn't it? If your answer is still 1. rather than 2., how can you justify that? Will I find the answer at WND? :D
The WND link can explain it better than I can. Do not be afraid. You have clicked on links before.
DeleteVincent, thanks for your respond here:
ReplyDelete"It should have been invested like a normal pension fund. Some money in bonds, some in stocks, some in real estate, some in gold. Investing at 0.25% just means they are getting further and further behind each year."
Well, is it really 0.25% they're getting? That's what they'd get if they had the funds in a reserve account, but they don't have access to reserve accounts, just plain old Fed deposits. I don't know the answer... do the nonmarketable bonds SS buys pay 0.25% a year? Do they have longer maturity bonds available that pay a bit more? I have no idea how it works.
It's a little weird to me to think of the gov agencies like SS investing in anything except the safest assets: aren't heads going to roll if they gamble and lose on the stock market? Who's going to want to take that gamble politically?
Not only that, but if they buy T-bonds, and then Tsy spends the proceeds (even if just on paying off principal amounts on publicly held regular T-bonds) then those $ are going back into the private sector where private sector actors decide how to invest them and decide what level of risk they should take. Is it really up to the Fed gov agencies to take those risks?
Maybe the answer is "yes" since state and local gov pension funds take those kinds of risks... but sometimes they get in trouble too.
And of course you can argue that if the bulk of the proceeds of nonmarketable bond sales don't go to pay off principals of marketable bonds, but instead are used for deficit spending, then the gov is again deciding how to invest those $... and I'd say that's true, at least initially. Whoever is the recipient of those $ though will turn around and be responsible for the risk decisions in investing them. Or a fraction of them anyway (assuming some dollars are not used as long term "investments" but rather for consumption).
Plus, perhaps it's the case that fed-gov trust funds (like SS) can afford to have super safe investments that don't quite keep up with the rate of inflation. There's probably some critical variables there: perhaps some rule of thumb such as as long as NGDP growth makes up for the difference, then current gov revenues should make up the difference. I'm not proposing that's an actual rule of thumb, just that maybe there's one along those lines. And notice I wrote NGDP, not GDP.
I guess I thought that from the start SS was intended as a way for current revenue streams from current workers to provide something for current retirees... not so much for people's individual gov-directed "investments" to grow and keep up with inflation.
OK Vincent, I went to the wnd site and read this:
ReplyDeletehttp://www.wnd.com/2011/03/278017/
But that still didn't clear much up for me. Can you explain this GAAP accouting and why you think that's the way to do the accounting? Is that what the WSJ article authors used? Thanks. I'd like to know the math here.
What do you think is meant by "present value of projected X over the next 75 years" where X could be GDP or a revenue shortfall for a program like SS, etc?
In other words, the WND article seemed pretty much like just another version of the WSJ article, except that they used the phrase "GAAP accounting." So why doesn't Sadowski's criticism (that I quote above) of the WSJ piece apply to the WND piece too?
Delete... also, do you think this is a case of being able to obtain any answer you want simply by picking the right year to project forward from (pick a year with a shortfall) and then projecting forward X many years to get the answer you want? For example, say I wanted the answer to be that we are $10T in the hole each year, then given a year with a budget shortfall, is it then possible to solve for X to get the $10T I wanted? Perhaps the answer is that I need to use at least a 300 year projection (i.e. X = 300). Do you suppose that's what could be going on?
DeleteClearly there are a lot of assumptions in anything 75 years out. What will the growth rate be? What will the inflation rate be? What will interest rates be? But there is a real problem. If a private company had a pension fund that was getting further behind so badly each year it would probably be shut down for bankruptcy.
DeleteHere's my suspicion written out using "maths." Say that the growth rate in the revenue shortfall for SS, Medicare, and other liabilities is, this year S%, and the inflation rate is I%. And say that S > I. And say the current year's shortfall was $K. Then we could attempt to project forward Y years, while simultaneously accounting for inflation like this:
DeleteT = total shortfall over Y years in today dollars = sum(i = 1 to Y, K*(1+(S-I)/100)^i)
Well that's a summation of a growing exponential, and it's guaranteed to produce for you an arbitrarily scary number, depending on how big you make Y, even if you "normalize" by Y at the end. In other words, we can make the ratio:
T/Y = the amount we need to pay each year in today's dollars
as frightening as we want simply by making Y large enough. Do you think that's the kind of game that's being played here?
... and the irony is if you'd selected a year with S < I to project forward from, then you could select Y such that
DeleteT/Y
was as arbitrarily close to zero as you wanted! Ha!