September 12, 2010
Should the Fed try to depress long-term yields further?
I've been sharing with readers my recent research with Cynthia Wu, in which we found that the Fed could likely lower long-term interest rates further by buying more long-term securities, even though the short-term rate is essentially zero and even though the newly created reserves would simply sit idle in banks' accounts with the Fed. Here I'd like to take up the question of whether such a policy would be desirable.
According to the framework proposed in our paper, the essential characteristic that allows nonstandard open market operations to be able to influence relative yields is the fact that long-term government debt has different risk characteristics from short-term debt. If you buy a one-year Treasury bill today, you know exactly how much money you're going to have one year from now. On the other hand, if you buy a 10-year bond and sell it one year from now, you will have a capital gain if interest rates go down but a capital loss if interest rates go up. Because nobody knows for sure which is going to happen, you're exposed to a risk with the longer term bond that you can avoid with the 1-year bill. Historically, holders of longer term securities received a higher yield on average as compensation for that risk.
If the Fed were to buy a large enough volume of long-term debt, it might be able to reduce the net risk exposure of the private sector so as to change slightly that average compensation and flatten the slope of the yield curve. Whether that's indeed possible, and how big a change in rates we might expect to see, is an empirical question. What Cynthia and I found, based on what was observed on average over 1990-2007 in response to modest changes in the maturity composition of publicly held Treasury debt, is that replacing $400 billion in outstanding long-term Treasury debt with short-term debt might lower the 10-year yield by 14 basis points. Our estimates also imply that, in the current environment when short-term rates are essentially zero, if the Fed were to buy about $400 billion in long-term Treasury debt outright with reserves newly created for that purpose, it might still be able to reduce the 10-year yield by about 14 basis points.
The first point I'd like to emphasize is that, although I described the operation above as something implemented by the Federal Reserve, it's really much more natural to think of as something for the Treasury to do. All the Treasury has to do is stop issuing debt of maturity longer than one year, and do its new borrowing with short-term T-bills, to accomplish the same thing. That would lower the Treasury's borrowing costs and stimulate the economy. So why wouldn't the Treasury want to do such a wonderful thing?
The answer seems pretty obvious. The reason the Treasury has never relied exclusively on short-term borrowing, and the reason it wouldn't be willing to today either, is because if all its debt were short term and interest rates subsequently go up, the Treasury could be faced with a ballooning refinancing problem that could be very hard for it to fill. The Treasury doesn't want to face the risk of higher interest rates any more than private investors do. If the Treasury were more willing than private investors to absorb this risk, then yes, by doing so it might be able to lower long-term yields. But as a practical matter, the Treasury may be quite unwilling to absorb this extra risk.
And if we suppose that the Treasury has good reasons to try to avoid exposure to this risk, what sense could it make for the Fed to absorb the risk on behalf of the Treasury? I find it hard to come up with an answer to that question. The most natural perspective for me is that the Fed should be more risk averse rather than less risk averse than the Treasury.
It is true that price stability-- avoiding excessive inflation or deflation-- has traditionally and quite appropriately been regarded as the responsibility of the Fed rather than the Treasury. But I think the most important tools at the moment to prevent deflation would be exchange-rate targeting, fiscal stimulus, and direct credit extension, along with the possibility I'm discussing here of changing the maturity structure of publicly held government debt. And all of these would more naturally be directed by the White House rather than the Federal Reserve.
The one advantage I can see for the Fed in this regard is a political one. If the Treasury were to stop issuing long-term debt, it would lead to open public debate about whether this course represents prudent management of fiscal risk. If the Fed does what amounts to the same thing via open-market operations, the details are sufficiently arcane, and the Fed is a sufficiently independent institution, that directly elected public officials would not be the target of criticism for such operations. This in fact seems to have been an important reason why the Fed and not the Treasury was the chief actor in many of the initial policy responses to the financial crisis in the fall of 2008, even though those operations also had an essential fiscal as opposed to monetary component.
But that feature is precisely what makes me prefer to see such operations implemented by the Treasury rather than the Fed. I am a strong believer in democracy, warts and all, and am very leery of any effort to circumvent the process of legitimate democratic review and debate by those who fear the cumbersome process would derail their own superior designs.
In addition, there is the separate question of whether the exposure to additional interest-rate risk by the unified Fed/Treasury balance sheet would indeed continue to have the stabilizing effects suggested by the historical correlations if pushed to a starkly higher degree than anything in historical experience. The empirical estimates in our paper simply took the risk characteristics of long-term bonds as given by their historically observed behavior, and assumed those wouldn't change as the composition of publicly held debt changed. But it is certainly plausible that big changes in the Treasury's or the Fed's maturity profile might be regarded by some bond investors as themselves a new source of risk, separate from that in the observed historical data.
Here's where I come down. If we do see more disinflation and further deterioration of economic conditions, I think the Fed has no choice but to try to use its powers to pursue its mandate of promoting maximal employment and stable prices. My guess is that additional purchases by the Fed of long-term assets will be a necessary part of that.
But even though the Fed still has some ammo left, this particular gun is one that I believe they should fire with reluctance.
Posted by James Hamilton at September 12, 2010 06:56 AMdigg this | reddit
The question of whether depressing long-term yields is a fiscal or monetary policy depends largely on the mechanism used to implement it.
If the Fed were to shift the yield curve by selling short-term treasuries to buy long-term treasuries, that sort of zero-sum shift would affect the duration of the government's liabilities, representing the Fed influencing fiscal policy.
If, however, the Fed were to purchase long-term treasuries with newly created reserves, then they're just printing money with the explicit objective of stabilizing prices and warding off inflation. In other words, the Fed in this case is doing exactly what it ought to be doing.
I agree the Fed shouldn't tinker with the Gov's duration, but with the persistence of deflationary concerns, I would be happy to see "helicopter" Ben would start living up to his pejorative nickname. As for those concerned about hyperinflation, those issues are entirely driven by an unsustainable fiscal policy and, in my view, the Fed has no power to diffuse that issue.
Posted by: Tudor at September 12, 2010 07:49 AM
Tudor: I see newly created reserves as essentially equivalent from the perspective of the unified Fed/Treasury balance sheet as newly created 4-week T-bills.
Posted by: JDH at September 12, 2010 08:11 AM
Prof. Hamilton --
You say when a private investor buys a 10-year Treasury, they are exposed to the risk that yields rise and they suffer a capital loss after one year. Fair enough.
And you say that when the Treasury issues shorter-term debt, they are exposed to the risk that yields rise and they have to pay more to roll the debt. Also fair.
But when the Fed prints money to buy 10-year Treasuries, it is exposed to the risk that... What, exactly? That it might suffer a mark-to-market loss after one year? I am pretty sure the Fed does not look at it that way at all...
Posted by: Nemo at September 12, 2010 08:14 AM
Nemo:The risk to the Fed is if banks decide they want to do something with their trillion dollars in excess reserves other than hold them overnight for 0.25% interest, and the Fed cannot withdraw those reserves because the value of its assets has declined due to capital losses on its long-term bonds.
Posted by: JDH at September 12, 2010 08:31 AM
An interesting post, but it is unfortunate that you have once again make a totally flawed and misguided analogy between the Fed/Treasury and private households. You postulate without any kind of evidence that there is "risk" for the U.S. Government to issue short term debt only just like there would be for private investors.
Do you know of any private investors that are able to dictate the interest rate they pay on their own short term debt whatever the circumstances? Do you know of any private investors that are able to create money out of thin air so that a liquidity or solvency crisis is never an issue?
A country with its own floating non-convertible currency can never face a liquidity or solvency crisis. When the U.S. Government issues bonds, all its does is replace reserves with bonds. Issuing bonds for a country with its floating currency is a reserve drain operation. That's it, that's all. The U.S. Government may well choose to issue 3-month T-Bills only or could choose not to issue bonds at all and let excess reserves grow at the Fed. There would be no liquidity, solvency or other so-called fiscal risks resulting from these operations.
Posted by: Qc at September 12, 2010 08:51 AM
economists' thinking never ceases to amaze me and sometimes to outright amuse me. it is really the economics of ideas that does not cut it, although most suggestions are rich on intellectual musings.
why is the lowering of long term interest rates by the Fed a fruitless policy? because the very same reason you are pointing the Treasury does not borrow only short term: interest rates will inevitably rise. you have the option to lock your money at the supressed yield for 10 years and pick a couple of basis points more prior to an intervention for the entire period or wait 2-3 years and pick a couple of percentage points more when economic conditions improve. the interest you forego for 2-3 years will be less than the interest differential for the remainder of the 10 year period. for30 years, the difference is even more striking.
its the economics, economists!
Posted by: baychev at September 12, 2010 09:05 AM
If past records are indicative of future performances.
It will then be, difficult to assess the future impact of an additional marginal QE on the investment side of the banks ledgers, save the fact it may be be cheaper and available resource in order to nurture existing assets.
Series: INVEST, Total Investments at All Commercial Banks
The benefits of already depressed interest rates, short, MLT have not been translated fully, in private fixed investment either.
Series: FPI, Fixed Private Investment
Above graphics cast doubts on the ability of the Fed or CBs to be a full employment provider through QE and depressed interest rates.MLT only.
The issues are and remain assets prices,capital raising and assets disposal for the leveraged side of the financial industries.
An accident in the debts markets does not require to be a direct impact on sovereign debts issuers only,it may start with the solvency of a large private sector bond issuer and they are many of them.
Posted by: ppcm at September 12, 2010 09:49 AM
The more reluctant the Fed is perceived to be to fire that gun, the more likely it is they will need to fire it.
Posted by: Nick Rowe at September 12, 2010 11:00 AM
Qc: Your position that there is no such thing as a government budget constraint remains incoherent to me, though I see you have lost none of your self-assurance as to its veritude.
If there is no government budget constraint, then I don't know why I'm having to pay all these taxes, which I truly find most bothersome to pay. The last time we discussed your ideas, you seemed to agree that there was some sort of government budget constraint after all to the extent that, if taxes were eliminated all the way to zero, there would be a significant fiscal crisis and hyperinflation after all. I gather then that your worldview is, there is no budget constraint as long as the government collects some minimal amount x in taxes every period and spends no more than y. Please tell me how low we can choose x, given y, and still avoid any concern. Because I am looking for a numerical answer from you, please tell me the specific value of x when y is set permanently at the level of 20% of GDP.
Posted by: JDH at September 12, 2010 11:10 AM
The Fed can always reduce the excess reserves banks have by increasing the reserve requirement. This is a very blunt tool but may be appropriate in the face of Federal Reserve capital losses.
Posted by: Rajesh at September 12, 2010 12:20 PM
"but to try to use its powers to pursue its mandate of promoting maximal employment and stable prices"
imo, the fed board cares far more about stable prices (e.g. the valuation of bonds and securities owned by the rentier class) than unemployment.
10% unemployment should be causing red alert klaxons at 20th and constitution. instead we have darwinist rationalization about structural unemployment and calls to raise interest rates.
Posted by: yuan at September 12, 2010 12:27 PM
It's my fault, I think. If it's the same person, about 18 months ago I recommended Qc read Abba Lerner on Functional Finance. I now recommend Milton Friedman, to get a fuller perspective.
If currency is (say) 10% of annual nominal GDP on average in the long run, and if target inflation plus long run real growth mean that nominal GDP grows at (say) 5%, then the revenue from printing money is 0.5% of GDP on average in the long run, so y-x=0.5% of GDP. (It's about half that in Canada, because foreigners don't hold our currency, so the Bank of Canada earns about 0.2% of GDP profits from printing money, which helps finance a bit of government spending, but not much.)
Back on topic: the Fed's buying government bonds is problematic, because if the policy succeeds, real bond prices will probably fall, so the Fed makes a loss, and has less (or negative) profits to give to the government. The solution to this problem is for the Fed to buy something which rises in value if the policy succeeds. I suggest the Fed buys the S&P500 index. That will probably go up in value of the Fed can make the economy recover.
Posted by: Nick Rowe at September 12, 2010 12:54 PM
The debt-deflationary depression will result in bank loans contracting 40% from the peak and bank loans at 30% of GDP and 1:1 to the monetary base, coinciding with the Fed doubling the monetary base and banks increasing their holdings of Treasuries and agencies by a proportional amount to the decline in loans.
The Fed will move out the curve because banks (owners of the Fed) will direct the Fed to do so to (1) assist in the secular liquification of banks' balance sheets and (2) assist gov't borrowing and spending of $1T+/year or 9-10% of GDP over the course of the decade ahead. By the latter part of the decade, the gov't will have borrowed and spent an equivalent to 100% of today's private nominal GDP.
M2+ (M2 + institutional MMFs + large time deposits), if you will, is now 83% of nominal GDP (velocity of 1.19) and ~132-133% of private nominal GDP (velocity of ~0.75). With bank loan growth still contracting, net incremental federal gov't borrowing and spending is the only thing keeping money supply growing at present.
As private economic activity continues to contract from the secular debt-deflationary effects on private bank lending, federal gov't spending will continue at 7-10%/year just to keep nominal GDP from a sustained contraction.
As in Japan since the late '90s (when Japan's persistent debt and price deflation commenced), US M1 could grow by double digits as the Fed prints and buys Treasuries the gov't issues; however, M2+ will likely grow only as fast as gov't spending and the net effects of ongoing liquidation of stocks, real estate, corporate bonds, and corporate assets, which is likely to be at a rate of no more than in the ~2%.
M2+ velocity will continue declining with the money multiplier, with velocity to private nominal GDP well below the level at which new public and private debt will result in sustained private economic activity.
At the M2+ velocity to private nominal GDP, additional gov't deficit spending will contribute very little to private economic activity and further reduce M2+ velocity to private GDP.
Posted by: Nemesis at September 12, 2010 02:11 PM
I am not sure I am seeing a fall in economic conditions right now. If anything, the economy looks to have reheated during the 3rd quarter a bit. Most "indices" are lagging and that is giving false negatives.
Posted by: The Rage at September 12, 2010 02:24 PM
"The risk to the Fed is if banks decide they want to do something with their trillion dollars in excess reserves other than hold them overnight for 0.25% interest, and the Fed cannot withdraw those reserves because the value of its assets has declined due to capital losses on its long-term bonds."
What about raising the interest paid on the reserves? Is it possible that the required rate would be so high that they couldn't afford to do this? (If they printed money to pay the interest, it would foster the inflation they would be trying to stop, but would tis be limiting? (I am too lazy to work through the analysis, even with back-of-the envelope numbers).
Increasing reserve requirements, at least to any appreciable extent, seems a real non-starter, owing to different degrees of excess reserve holdings among banks.
Posted by: don at September 12, 2010 03:02 PM
Before responding to your question in a second post, let me say this... I know that my views sound iconoclast, weird, and waco, but I am just describing monetary and fiscal operations as they are run (I do, however, admit to being iconoclast enough to have a passion about the nitty gritty details of fiscal and monetary operations). Although my views sound weird for mainstream economists, you know just as well as I do that the facts are on my side :
1- How do you explain, according to your view of the world, that the yield on the US long bond is currently below 4% with these really "scary" deficits, while it reached 6% during the surplus Clinton years?
2- How do you explain that the most heavily indebted country in the world -Japan- is borrowing for 30 years right now at 2%, while one of the lowest indebted country in the world -Australia- is borrowing for 15 years at 5.06%?
3- Do you consider it a pure coincidence that the countries affected by the so-called sovereign debt crisis (PIGS) did not have their own floating non convertible currency?
4-How do you explain that countries with huge deficit and huge debt (e.g. UK, US, Japan, etc) that have their own floating non-convertible currency have actually seen the yields on their debt decrease during the so-called sovereign debt crisis in May?
5- When is the last time that a country with its own floating non-convertible currency suffered a liquidity crisis or defaulted on its debt denominated in its currency? (I was able to find one example, Japan that defaulted on its US creditors during WWII... of course this was a purely political decision)
6- How could we keep hearing these days that the level of the Japanese sovereign debt is really scary while at the same time hearing stories about the ever strenthening Yen?
7-When is QE going to generate huge inflation? (back when it started in the US, or in Japan 10 years ago, there was no shortage of economists to say that this will lead for sure to inflation very soon)
8-Do mainstream economists still think that lending excess reserves will make reserves disapear? (Nourriel Roubini just stated as much last week)
9-How come China is still piling on US treasuries despite repeted warning from mainstream economists?
10-How come private savings and fiscal deficit seem to evolve hand in hand?
I will stop here... I do find it interesting that some mainstream economists would, in the same sentence, declare that Japan is an unexplainable awkardity that cannot be generalised because of "cultural differences" and then go on to use Zimbabwe as a relevant analogy for the US or the UK.
Posted by: Qc at September 12, 2010 04:08 PM
I think I understand Qc's point. A govt that is sovereign with a floating fiat currency can create money. In fact, you can view all govt spending as "money creation" and taxes as "money destruction". As such, it can never be unable to pay its debts (assuming it only borrows in its own currency). There is no default risk, and no risk of a liquidity crisis.
Why tax? If government spending creates money, taxes destroy money. The reason to do so is to prevent government spending from being inflationary. Taxes destroy private sector spending power, so that government spending doesn't push aggregate demand beyond the economy's supply constraints, which would be inflationary.
Note that aggregate demand and supply capacity are both flow variables. If more "stuff" is demanded at prevailing prices than suppliers are able/willing to supply at those prices, prices will rise until someone is outbid. Stock variables like the size of the money supply don't enter into it.
As to the question, what level should taxes be to support govt spending at 20% of GDP, I'd say 20%, minus the desired inflation rate divided by the slope of the aggregate supply curve, minus the desired savings of the private and foreign sectors. The foreign sector has been a net saver for a while now, allowing the govt to run a substantial deficit without increasing inflation. Now the household sector wants to save, to spend less than their income. But all income is someone else's spending, so someone has to dis-save. Currently we're forcing 9.5% of households into involuntary dis-saving by denying them income, in order to make things balance.
If the amount that non-govt wants to hold as savings grows in proportion to the size of the dollar-denominated economy, then in the long run (averaged over multiple business cycles) the debt/GDP should stay about constant. But in the short run, the appropriate deficit and resulting debt level can swing quite a bit.
As to the fear that the Fed would find it hard to raise rates when the economy recovers (because they need to drain reserves but their long-term holdings have dropped in market value), simply have the Treasury issue new bonds to absorb the excess reserves, depositing the money in a special non-interest-bearing account with the Fed. Problem solved. Don's suggestion -- have the Fed pay a higher interest rate on excess reserves -- works too, though the Fed's balance sheet ceases to balance. There's still no solvency risk there, it would just be cleaner to keep the Fed's books balanced and all the debt in Treasury's name.
The overall concern about aggressive fiscal policy is a bit different: If the national debt grows a lot (say to 200% of GDP), and then the economy picks up and the Fed wants to raise rates, then the interest paid on the debt becomes a large enough flow variable (income to the private sector) boosting aggregate demand, that it might be inflationary. Here the answer would be to raise taxes -- reduce aggregate demand through fiscal policy, so that rates can remain lower.
Posted by: lilnev at September 12, 2010 05:56 PM
The Fed could commit to keeping one-year rates low for the next (say) 5 years, and therefore manage the risk of incurring a capital loss on long-term bonds in a way that Treasury could not. If so, might this be a good reason for the Fed, rather than Treasury, to purchase long-term bonds?
If the policy objective is just for the public sector to take on interest rate risk on long-term bonds, then perhaps it makes no difference whether Treasury or the Fed purchases the bonds. If the policy objective is for the Fed to signal its commitment to low short-term interest rates for the foreseeable future, then perhaps the Fed should purchase the bonds. Treasury could guarantee that it will compensate the Fed in full for any loss on long-term bonds.
Posted by: Kien at September 12, 2010 06:34 PM
The Fed doesn't have risk the same way the rest of us hapless folks with no printing press have real risk. Their risk is they lose control of shrinking the money supply and/or raising interest rates when the time comes if they have a bunch of long term bonds on the balance sheet that don't trade for what the Fed paid for them whenever the Fed needs to sell them. Their portfolio also needs to generate the interest necessary to pay interest on reserves to make Ben's new monetary idea work. They could just print the money to pay interest to the banks, but at some point the world will look askance at all this.
So that is the risk. When domestic and foreign workers, investors, suppliers and producers begin to feel the Reserve is paying everyone in Attaboys or Gold Stars, then the problems start.
It's happened in many Third World countries already, so it will only be a new thing when the World's Reserve Currency country does it.
Posted by: Cedric Regula at September 12, 2010 08:23 PM
"I will stop here... I do find it interesting that some mainstream economists would, in the same sentence, declare that Japan is an unexplainable awkardity"
I've heard Japan described as "A Bug Waiting for a Windshield." But it was a financial guy that said that.
Posted by: Cedric Regula at September 12, 2010 08:36 PM
A young William Jennings Bryan became the Obama around the turn of the 20th Century with his Cross of Gold speech at the Democratic convention, forcefully advocating for loose money. Before the creation of the Fed, monetary policy was a subject of public debate and elections could be won and lost over it.
Prof. Hamilton has alluded to the convenience of having these tough decisions taken out of the political realm but also to the risk. As a result of removing monetary policy from politics, the central bank has been able to take big risks without debate. Although the OECD and the BIS pointed to Basel and loose monetary policy as stimulating the asset run up, nary a word has been spoken in the elections at the turn of the 21st Century about this important subject.
In spite of Professor Hamilton's wish that this be done in the context of democratic debate, I do not think he should hold his breath waiting for it. Democratic debate exists only because politicians can be held accountable at election time for their decisions. When such decisions are made by a unelected technocratic elite no politician is in danger of losing his seat over the them. Obama the gifted young orator will not make any Cross of Gold speeches. Monetary policy will remain largely unscrutinized, and the debate will continue to rage over whether to crack eggs at the big end or the small end.
Posted by: colonelmoore at September 12, 2010 08:56 PM
Is there no concern that another blast of fresh money, applied this time to long-term debt, will distort relative prices?
Are correct relative prices not necessary to an efficient & sustainable economy? If not, then I guess it is fine for a gov't bureau to manipulate the most central of all prices all it wants.
Posted by: Bryce at September 12, 2010 09:06 PM
Back to your questionning, it is true that Modern Monetary Theory economists -clearly a minority of a minority of economists- consider preposterous any claims to the effect that a country with its own floating non convertible currency is revenue constraint from a fiscal standpoint and therefore should keep its debt-to-GDP below a given arbitrary level . This being said, this proposition does not change the fact that any country is constrained from a real ressources standpoint (you can call it the economy's supply constraint). Therefore, a country with its own floating non convertible currency is constrained not by the money it is able to "borrow" from the private economy but rather from the real ressources of the economy.
In such context, what is the purpose of taxation? First, as I explained in a previous post, taxation is a sufficient condition to create demand and value for a given currency. Second, the purpose of taxation in a country with own floating non convertible currency is to regulate aggregate demand, that is to regulate access to the real ressources of the economy. Regulating aggregate demand usually mean that some people in the economy will have to forgo access to some real ressources to make sure some other people (in the case of Govt transfer payment) or the government itself (in the case of Govt brick and motar spending) will be able to access these ressources. Regulating aggregate demand could also mean discouraging certain behaviour like tobacco consumption through higher sales tax.
Of cource, your question seems to be why should some people have to forgo access to real ressources as a result of an income tax if the government does not need this tax money to finance its spending. Well, it is true that a country with its own non convertible currency does not need this tax money to finance its spending, but it nevertheless still need some of its citizen to forgo their access to the real ressources of the economy so that some other people or the government itself will have access to these ressources. What would happen if government bring taxes down to zero while maitaining its spending, and therefore do not ask any of its citizen to forgo consumption of real ressources? A fight for access to real ressoures would likely result in the form of a price bidding war and inflation will result as the real ressources of the economy would not be sufficient to supply all citizens with what they are demanding. From a mainstream economic perspective, taxation is there because the US government needs your money to finance its spending. From the Modern Monetary Theory (MMT) perspective, taxation is there because the US Government needs you to forgo consumption in order to have other citizens or the government itself have access to the real ressources of the economy without generating a price bidding war. It is precisely to avoid this price bidding inflation that the US Government implemented quantity rationing during WWII, a time of huge fiscal deficits and full employment. Conversely, it is easy to see why in the current circumstances -with a lot of real ressources sitting idle in the real economy- MMTer have been huge supporter of a national payroll tax holiday and an increase in the fiscal deficit. In the current context, you can increase access to the real ressources of the economy to your citizen without having other citizens forgo their access, and without much inflation because of all the ressources sitting idle.
With government spending at 20% of GDP in the long run, what level of taxation would be needed to "avoid any concern"? Lets assume that current account is in balance. The national identity could then be expressed this way:
S-I ≡ G-T
If you fix G at 20%, then the desire to net save (ie. S-I) of the private economy in the long run will dictate what T should look like in the long run as well. If the desire to net save is zero, then T could roughly equalled 20%. If the private economy desire to net save is immense, then T would need to be much below 20%. Of course the question could then be, how do we know the desire of the private economy to net save? I would say that the unemployment rate is a very good indicator to start with.
Posted by: Qc at September 12, 2010 09:21 PM
I have a relative that works for a big Japanese bank and has access to the reports that bond traders and hedge funds all read.
He tells me that hedge funds are buying derivatives betting against the yen. The Japanese people have always been big savers and one of the largest ways they saved was through the Postal Savings Bank accounts. This bank is a big slush fund for political pet projects. Its primary purpose was to invest in construction boondoggles to keep the LDP in power but the government has also taken a portion of those deposits to buy JGBs. Thus the Japanese deficit was financed 93% by domestic buyers.
Now that the Democrats are in charge they have cut back on construction projects and passed a law allowing the government to take back the shares of the semi-privatized Postal Savings Bank and also allowing the banks to invest a larger percentage of deposits in JGBs. Since the government is the majority shareholder the bank does what its owner tells it to do. But that is not enough so the government has now started advertising in trains, saying that women think that men who buy JGBs are sexy. This reminds me of California's radio ads for its residents to buy its debt.
This is because the Japanese Baby Boomers, who are an average of 5 years older than ours, are retiring and becoming net withdrawers of savings. Furthermore the Japanese birthrate is 1.3 kids per couple and the population has started to decline, so there are simply not enough new savers to make up the loss. So it is only a matter of time before the lines of government borrowing and domestic savings intersect. Assuming the government doesn't confiscate private savings, it will then have to offer its debt to outside investors.
No rational investor would buy low-interest bonds of a country whose government is running up such a huge debt with a population that is retiring en masse and a collapsing taxpayer base. So when Japan hits the capital markets, interest rates will climb ajnd it will fall into a debt spiral.
These things take time but the collapse will absolutely happen. For now Japan has enough domestic deposits that the government can commandeer to keep the shell game going. Spain, Portugal and Ireland, on the other hand, depend on outside investors to stay afloat and on the hostile voters of Germany to put up with rescuing them if the investors balk. And Japanese don't firebomb foreign banks when their wages get cut. By comparison to Europe, Japan looks good in the short term. But when the punch runs out, the party will certainly be over.
I am certain you will not agree with me, but anyone can see where things are going using only grade school arithmetic. A amount of domestic savings, B percentage of savings put into JGBs, government debt rising at rate C with no end in sight, D net withdrawals.
Posted by: colonelmoore at September 12, 2010 10:04 PM
2- you are entirely missing the point: one thing is to borrow in the market, another thing is to borrow with a political arm twisting of private companies. the source of financing and opportunity cost matter as well quite a bit in your japan/australia comparison. another reason is the probability of default, that day of reckoning is being delayed in japan that does not rely on external financing. when the creditor is the debtor is the creditor and so on, things get very muddy.
3- the PIGS debts are no coincidence but the result of using the northerners' currency. if PIGS still had their own currency, they'd be borrowing not in it, but in foreign currency. having their own currency would not help them in their present situation, it would have helped by preventing them to become overindebted.
4- again you view it solely from the point of sovereign risk. do a relative comparison.
5- how many countries have a floating non-convertible currency and are able to borrow in it? does this explain the scarcity of such defaults happening?
6- the yen strengthening is a result of the trade balance and other minor factors while imposing restrictions on the ability of households to repatriate swiftly assets abroad effected this year.
7- interest question: when you choked people on debt and they can hardly service it, it is difficult to create inflation in levered assets even if you begin to monetize the debt. there is a clear relative repricing of goods though. cash purchased goods inflation rate is well over 100% in the past 10 years while levered assets are down in japan. hope you undestand where inflation and where deflation occurs.
9- china has no other option to keep its own people busy otherwise they will revolt. what would you rather do: lose some common money or lose your personal power?
Posted by: baychev at September 12, 2010 11:57 PM
MMTer's are nutcases.
Posted by: Cedric Regula at September 13, 2010 07:19 AM
"This is because the Japanese Baby Boomers, who are an average of 5 years older than ours, are retiring and becoming net withdrawers of savings".
If they withdraw their savings in Yen and spend them, where are these Yens going to go? I will tell you where they are going to go: in someone else pocket. These Yens will not disapear from the private economy and at the end of the day they are going to be used to buy Japanese Bonds. The act of withdrawing savings and spending them does not result in money disapearing from the private economy.
Posted by: Qc at September 13, 2010 07:22 AM
Thanks for the sound academic counter argument. I will take this important fact into account in shaping my understanding of fiscal and monetary operations.
Posted by: Qc at September 13, 2010 07:52 AM
QC, I don't feel like typing the bizillons of words it would take to counter the bizillions of words typed by MMTers only to redescribe how the world works as we know it instead of the MMT version which redescribes the world in there terms.
Posted by: Cedric Regula at September 13, 2010 08:19 AM
Ok, fine with me, don't tell me how the economic world works acording to your view. But please answer to following basic question on fiscal and monetary operations:
In a given day, the US Government generates a net cash inflow of 50 billion (Government tax receipt are above spending that day by 50 billion). The US Treasury decides not to repurchase its debt that day and simply leave the 50 billion in its account at the Fed. Could you please tell me what would happen to the Fed's balance sheet before the day is over?
Posted by: Qc at September 13, 2010 09:09 AM
The Fed and the Treasury differ in one very important respect: the Fed can, at any time, just stop paying interest on excess reserves. If the economy begins to recover and the Fed wants to reduce excess reserves, it can always use the blunt instrument of raising reserve requirements.
The Fed should not try to twist the yield curve. As you say, that's fiscal policy that is properly handled by elected officials. The Fed should stick to monetary policy, for which it does have constitutional authority delegated by the Congress.
What the Fed should do right now is (i) stop paying interest on excess reserves, and (ii) announce a price-level or NGDP target path. Open market operations in pursuit of the target should be neutral with respect to the maturity structure of Treasury debt. For two years now we have had, as Scott Sumner keeps pointing out, excessively tight money. Before you reply that money has not been tight, please read some of his posts on why expected NGDP is the correct measure of how tight or easy money is.
Posted by: Jeff at September 13, 2010 09:27 AM
BTW, the Japanese Boomers are 12-15 years older than US Boomers; therefore, the worst of our debt-deflationary depression with periodic or persistent consumer price deflation (or at least core prices) is implied to last from 2010-12 to 2020-22.
And it is during the same period that the worst 5- and 10-year returns for stocks will occur, as well as an eventual 45-50% decline in the median US house price (60-65%+ in the bubbliest areas) from the secular (permanent in inflation-adjusted terms) peak in '06-'07.
All gov't can do is borrow and spend at 10% of nominal GDP and 13-14% of private GDP to keep nominal GDP from contracting along the way. The average trend rate of nominal GDP will decelerate from 5% in '00 and less than 4% today to around ~1.7-2%, with private real GDP decelerating from the long-term 3.3% rate to a post-'00 trend rate of ~1% through the end of the decade.
With total gov't spending (56-57% of private nominal GDP) and private debt service at an equivalent of 80% of private nominal GDP, there can be no organic growth of private economic activity until gov't spending and private debt service costs either (1) "catch down" to wages/salaries and domestic production or (2) wages/salaries grow fast enough and for a sufficient period to catch up to gov't and debt costs. The clear implication of a secular debt-deflationary depression is that (1) is the outcome we face.
And for local and state gov't spending to "catch down" to the likely secular post-'00 trend rate of private economic activity, gov't spending and private debt service will have to be cut 30-35% by decade's end and as much as 45-50% by the mid- to late '20s.
Thus, anyone on this list who relies upon local and state salaries, benefits, and pension payouts (or hopes to) had better begin TODAY to adjust one's spending and lifestyle choices to adapt to one-third and then one-half of what one has been conditioned to expect to receive from gov't.
In effect, your private sector counterparts cannot afford what elected officials have promised you, and neither can most of you afford to pay higher contributions and out-of-pocket benefit costs to sustain the unsustainable commitments made with other people's money.
Posted by: Nemesis at September 13, 2010 10:34 AM
Agree that demographics seem to point to Japan finally cashing in on some of its foreign investments. However, the transfer problem would be huge. Their economy has been built on export-led growth, and a trade deficit would mean accepting a decline in AD that would depress the local economy. And the transfer would require a gain in the yen, not a fall. We might be seeing this happening right now.
A dramatic fall in the yen would be a boon to its exports and spur the local economy, but would mean an increase, not a decline, in net foreign lending. Nor can I see how a problem in meeting domestic debt would cause a fall in the yen - the debt is entirely internal, Japan being a huge net creditor internationally. An attempt, for example, to cash in some of its huge foreign reserves would cause the yen to rise, not fall. In fact, I think Japanese saving is an important contributor to low interest rates in the ROW, including the U.S. I would be more worried about the effect of Japan becoming a net dissaver on debt costs in the U.S. and U.K. (The local interest rate on Japan's debt would be kept low by interest arbitrage and yen appreciation.)
Under current circumstances, AD is scarce globally, and in my view will remain so for a long time to come - perhaps more than a decade. After all, it has been a decade since the dotcom bust, and in my view, global AD has been scarce since that time - the fact was just disguised by housing bubbles, caused in no small part by a savings glut abroad, particularly in Asia. These countries adopted Japan's example and built up something like $5 trillion in foreign reserves since 2000, which amount to forcing savings on the ROW.
Hedge funds have made some big mistakes in the past. I recall some big U.S. players touting Japan's stock market some years ago when the NIKKEI was at 11,000. But in this case, they don't even seem to have a coherent story.
Posted by: don at September 13, 2010 10:45 AM
I did read your question a few times, and the only response I can come up with is:
1) Who cares about one day?
2) What planet have you been observing economics and finance on?
Posted by: Cedric Regula at September 13, 2010 11:46 AM
my answer to your question :
Timmy asks Ben, if its allright to withdraw the 50 billion after 6pm and put it in Lloyd C. Blankfeins underpants.
yes, from 2010-12 till 2020-22 everything looks really dark and maybe we will all gonna die.
Yes, time to buy stocks now. Agree with that.
Posted by: Johannes at September 13, 2010 12:37 PM
Johannes, if you perceived my comments as bullish for stocks, you had better read the post again, and this time with comprehension.
When the S&P 500 falls to the 400s-500s and below, and the 10-year Treasury yield is around 1.5%, check back with me and we can talk about perhaps buying stocks; until then, run far away from "scams", a.k.a. stocks.
Posted by: Nemesis at September 13, 2010 02:44 PM
Qc at September 12, 2010 09:21 PM: OK, let's say government spending is permanently 20% of GDP, taxes are permanently zero, private saving is 20% of GDP, and the interest rate is zero. Then each year, those buying the Treasury debt increase their holdings of that debt by 20% of GDP. In 20 years, their paper wealth will have gone from nothing to 400% of GDP. There has been nothing left for investment and no economic growth.
Could this be an equilibrium, with debt growing to an arbitrarily large multiple of GDP? The answer is unambiguously no. Why would I, as a consumer, watch myself become a multi-millionaire (that government debt counts as my personal asset) but never spend a dime more than I do right now?
These intertemporal consequences of debt accumulation for the willingness of lenders to continue to buy government debt are the essence of the government budget constraint. A model that ignores those consequences makes no sense to me.
Posted by: JDH at September 13, 2010 03:17 PM
Qc: I have debated MMTers many times, and wouldn't mind doing it again, if I thought it would get us anywhere. But not here. James Hamilton has written a serious post, based on some serious research. Yes, he does not share your assumptions (about the LRAS curve being horizontal). But this post is not about MMT. So let us leave it, and stop taking up his space any more.
Posted by: Nick Rowe at September 13, 2010 03:27 PM
The goal of my question was not to embarrass you. I just wanted to show you that you may be able to learn something from the "MMTer nutcases" as you call them.
I did not find any economists able to respond to this kind of basic question on fiscal and monetary operations other than MMTers. I do respect them a lot for that.
Posted by: Qc at September 13, 2010 03:53 PM
Borrowing money not needed for the USA. It has been done in the past.
Posted by: Adam at September 13, 2010 04:08 PM
MMT does not ignore these consequences. The day that the private economy decide to put its massive accumulated savings to work -and that the productive capacity of the economy is still intact- will precisely correspond to the day when the private economy will start roaring again on its own, the economy will converge toward full employment and the government will no longer have to run deficits (in fact it would likely end up running a surplus thanks to the automatic stabilisers). The debt-to-GDP would then start shrinking. Note: Of course the current account balance somewhat complicate this picture, but lets assume it would be still at zero.
So I certainely do not think that a very large multiple of GDP in the form of public debt represent an equilibrium. At the same time, I am appalled by the definition used by the mainstream economic framework which states without any kind of theoretical or empirical justifications that "the net present value of all future primary balances must be sufficient to pay back the initial debt". See http://www.imf.org/external/pubs/ft/tnm/2010/tnm1002.pdf
Posted by: Qc at September 13, 2010 05:57 PM
Ditto. These followers of MMT remind me of the bald headed guys wearing gowns that used to hang around at airports. Always willing to share some enlightenment you didn't need. Except they hang around econ blogs.
As far as I can tell, the Weimar Republic invented MMT and these guys just stole it. They should at least call themselves Neo-Weimaraners.
Posted by: Cedric Regula at September 13, 2010 06:38 PM
This thread was about Prof. Hamilton's research so I won't go on beyond this comment. Your assumption is that people that withdraw savings will spend it and that this will eventually go into JGBs. But why must it happen? The government does not force people to buy its debt. They buy it because they are depositing their money into the Postal accounts, which the government does force to buy its debt.
I had Japanese friends call me in the height of their banking crisis in the first half of the 2000s to ask how they could withdraw their money and move it into US dollar accounts. I told them that our banks were in much worse shape than theirs and to sit tight. (Yes I do blush when people call me a prescient financial wizard.)
So why wouldn't the Japanese become equally fearful that their government is insolvent and start cutting back on their purchases of JGBs?
You made some leap of logic when you said that the money that is withdrawn at the end of the day goes into JGBs. Give me a concrete example of how that works. Let's take the cash someone spends on new shoes or a car and trace it through the economy until it becomes government debt. (Your assumption is that it becomes government debt at the same low interest rate because the government doesn't have to beg borrow and steal to get it.)
Don you said that you agreed that demographics would require Japan to cash in some of its foreign investments. You can think that but you can't agree with me because I never said that. I said that the government (not Japan which is composed of private entities as well) will have decreased tax revenues, and so it will have to finance its obligations with more borrowing. Because people are becoming net dissavers, this will have to come from foreign buyers of JGBs. And they won't want to lend for the same low interest rates in such a situation. Then Japan's government will be in a debt spiral.
Posted by: colonelmoore at September 13, 2010 10:53 PM
@ don at September 13, 2010 10:45 AM
don, your analysis is sound, but you disregard the fact that it is not in a country's interest to have its tax collections collapse, just to permit pensioners to repatriate foreign assets and increase their relative wealth. how would you finance the budget deficit? by imposing a windfall profit tax?
if pensioners want to bring back home their foreign savings, the gov't will print new yen to satisfy them and encourage workers to export their earnings to protect their value. this is the most commonsensical solution.
politics is about keeping the status quo, the best reference is the 'a change you can believe in' president ;)
Posted by: baychev at September 13, 2010 11:49 PM
Cedric: "As far as I can tell, the Weimar Republic invented MMT and these guys just stole it. They should at least call themselves Neo-Weimaraners."
There's more truth in that than you may realise (or perhaps you already knew this). Knapp's "State theory of Money" was a big influence on both.
Posted by: Nick Rowe at September 14, 2010 05:25 AM
Interesting. Didn't know that. But just in giving MMT the sniff test, it seemed to ring a bell.
Posted by: Cedric Regula at September 14, 2010 06:20 AM
Actually, as distasteful as it is, the effective use of fiscal and monetary policy under a floating exchange rate expressed by MMT is better reflected by the low inflation, high deficit and high growth period of the 1930s Germany, not Weimar Germany.
As a matter of fact, your Weimar argument rather than destroying the argument of MMTers actually supports it. They argue that a drop in productive capacity (from the war) and forced foreign currency-dominated debt are a recipee for disaster and hyper-inflation. It is the lack of sovereignty and the war reparations that cause the inflation.
Speaking of 1930s Germany. There is a lot of name-calling and insulting on this board. How many of you are familiar with Godwin's Law (http://en.wikipedia.org/wiki/Godwin's_law). The loser is the person who resorts to name-calling and hyperbolic comparisons first.
On economic blogs, you always get those hyberbolic comparisons to Weimar, Zimbabwe and the Soviet Union. We are increasingly getting farther away from healthy levels of inflation (deflation) in OECD countries that Weimar should be the last thing on our minds.
People, where is the excess demand? Where is the inflation? We have so much excess capacity right now, it is incredible. Unemployment is at 10% and output is way below trend. Forget MMT, how about Richard Koo? Balance sheet recessions, Hello. Financial Crisis, anyone? Even Krugman thinks we need stimulus and that guy is a borderline monetarist.
Posted by: BD at September 14, 2010 07:00 AM
Thanks both of you for your helpfull academic points.
I do not intend to go beyond this comment either. But let me respond to your question. You have to know how banking operations work on a day-to-day basis to understand this statement -and this is precisely where I think MMT is usefull. Government bond is the instrument that is used to allow the interbank market to clear (i.e. to reach an end-of-day settlement balance of close to zero). End of day settlement balance of close to zero means having no money sloshing around the interbank market since this "excess money" would tend to push the interbank market interest rate below the target as set by the Central Bank (this is particularly true if no interest is paid on deposits at the Central Bank). It is precisely the role of the Central Bank, through its supervision/management of the interbank market, to make sure that this market clear day after day at the overnight interest rate it dictates. For this to happen, the Central Bank stands ready to buy/sell Government bonds.
I will give you a concrete example as you asked for. In a case where Japanese sell Government bonds and withdraw their savings from Bank A to buy goods from a manufacturer that has a saving account at Bank B, the Central Bank would make sure that Government bonds are transfer from Bank A to Bank B so that the interbank market clear and the target for the overnight rate is achieved. You would note that the Central Bank does not force Bank B to buy the Government bonds, but as a good profit maximiser, Bank B much prefer to have some kind of interest on its Yen (even if very low) than stacking Yen under its mattress at 0% interest rate. And from a risk perpective, there is no difference for a private bank between holding Japanese Government bonds or holding Yen. This is precisely the problem with Greece and other countries that do not have their own convertible floating currency: there is a big difference between holding Greek bonds and holding Euros. Some european banks currenctly prefer to park their Euros at the ECB at 0.25% instead of investing in Greek bonds at a yield of 10%!
Posted by: Qc at September 14, 2010 07:48 AM
So...I guess you are saying that MMTers are more like Nazis than Neo-Weimaraners?
Here's a little refresher on the 1920s and 1930s in Germany.
And yes, "Hello, we have a balance sheet recession."
Myself and some others I recognize here from other blogs have been saying we are going to have that ever since around 2006. We didn't know it was called "balance sheet recession" at that time tho. So why does this this Koo guy get all the credit for being an economic genius?
Posted by: Cedric Regula at September 14, 2010 07:55 AM
Thanks for the refresher:
"But the "London ultimatum" in May 1921 demanded reparations in gold or foreign currency to be paid in annual installments of 2,000,000,000 (2 billion) goldmarks plus 26 percent of the value of Germany's exports. The first payment was paid when due in August 1921. That was the beginning of an increasingly rapid devaluation of the Mark which fell to less than one third of a cent by November 1921 (approx. 330 Marks per US Dollar)."
Thanks a lot. That confirmed what I already knew and argued. Are you trying to prove my point or yours. Just so you know, it was not Weimar policy to pay these reparations, they were forced to pay them by the victors of WW I.
You can name-call as much as you like. All I know is that I have not accused anyone of being a Nazi.
By the way, just so you know the solution for debt deleveraging, "Balance Sheet Recessions" are bigger deficits. Have you been suggesting that solution since 2008, as well. Our do you think that the US will have problems rolling over its debt. Do you think excess reserves are inflationary?
Here is a question for everyone, what should a central bank do to generate inflation? The issue that no one is addressing is that central banks couldn't generate inflation even if they wanted to. Don't name-call just explain in simple language the steps and actions that the central bank (such as the Federal Reserve) would take to generate either modest or hyperinflation?
Posted by: BD at September 14, 2010 09:18 AM
This seems earmarked to become one of those long MMT "debates" (See QC attempt above at FOMC ops. See St Louis Fed website for answer on that. ALSO DO NOT become infatuated with banking clerical functions that occur over a 24 hr period and try to conflate that to the next General Theory of Economics. If you want to know what's happening in Japan, read colonelmoore, baychev and don)
And I don't think anyone wants to do that.
so I'll just wrap up.
"Thanks a lot. That confirmed what I already knew and argued. Are you trying to prove my point or yours. Just so you know, it was not Weimar policy to pay these reparations, they were forced to pay them by the victors of WW I."
A: Let's just call it government debt.
"You can name-call as much as you like. All I know is that I have not accused anyone of being a Nazi."
A: Your Leader will not appear until another 5-10 years.
"By the way, just so you know the solution for debt deleveraging, "Balance Sheet Recessions" are bigger deficits. Have you been suggesting that solution since 2008, as well. Our do you think that the US will have problems rolling over its debt.
A: That's your solution to a problem that can't be solved, no, yes.
Do you think excess reserves are inflationary?
A: yes, no, maybe, depends. plus we would need to define "inflation".
Here is a question for everyone, what should a central bank do to generate inflation? The issue that no one is addressing is that central banks couldn't generate inflation even if they wanted to. Don't name-call just explain in simple language the steps and actions that the central bank (such as the Federal Reserve) would take to generate either modest or hyperinflation?
A: Central Banks shouldn't try to generate inflation, especially since they have such a spotty view of what inflation is. One very good reason not to try is we already know they can't seem to "generate inflation" in a liquidity trap.
Neo-Weimaraner, the choices the central bank have at this point is take the inflation rate that the economy gives us. Or destroy confidence in the financial system and then we get hyper-inflation. The knob on the printing press doesn't have a "moderate" setting.
Posted by: Cedric Regula at September 14, 2010 10:00 AM
I don't know why we are not even looking at our own country during WWII.
Massive deficit spending yet no inflation and no increasing interest payment burden. How was it done?
You guys are the economists, figure it out.
Posted by: Adam at September 14, 2010 10:12 AM
Okay - How about making MMT simpler? Since MMT just explains.
Treasury borrows from Fed. Fed sells Treasuries.
That is all.
Sure there are fancy smanshy short term and long term treasuries that I don't know about.
Fed sells Treasuries is the point of this post.
But from MMT explanations. It is not needed to finance government borrowing. Only Treasury borrowing from Fed is needed. (Since that is what is in the Constitution)
So the Fed and Treasury could induce inflation by stop selling Treasuries. Then they could disinflate again by selling Treasuries.
Simple even a Cave Man could do it. hehe
Then there is the monetary tools. I think we got those to where we can't use them right now.
Posted by: Adam at September 14, 2010 10:29 AM
The printing press has nothing to do with the way that the federal government spends. Treasury credits accounts. Money appears as deposits in bank accounts, there is very little cash per se in the system. Printing money refers to the breaking of a gold standard or currency peg.
With respect to the US Federal Reserve, there is no such thing as printing money (that option does not exist in institutional reality). Printing money is not a tool that is available to the Fed. Printing money is like pigs flying in the US. If by printing money you mean to say "Do not pay interest on excess reserves" than I disagree (why is 0% on excess reserves more inflationary than 0.25% - the interest rate currently paid on excess reserves). Moreover, for you policy wonks out there, helicopter money is fiscal policy, not monetary policy.
Adam: I disagree with your statement. Excess reserves and t-bills are almost perfect substitutes. Neither is inflationary.
Deficits can be inflationary, but not with such high levels of excess capacity. Payroll tax holiday for everyone.
Posted by: BD at September 14, 2010 11:01 AM
The extent to which the federal reserve would generate revenue on its purchases (by buying the S&P 500 index) is a tax on the private sector. Any revenues remitted to the treasury are a tax and they increase leverage in the private sector, destroy money and decrease aggregate demand.
Second, nobody believes that the LRAS is horizontal. That said, the current level of excess supply is grotesque. Growth rates were higher when unemployment was lower in the immediate post war era. Why grow at 2% when you can grow at 3%. As you mentioned Say's Law is garbage. If so, the government can have an impact on output. Let's stop focusing on what the government shouldn't do.
Lastly, in response to Professor Hamilton's blog posting. The federal reserve can control the whole yield curve for government securities. Refinancing has never been an issue for a country with its own currency and nobody has proven otherwise. Interest rates are set politically and not in the marketplace (unless the central bank chooses to let be set in the marketplace, such as long-term yields). This whole discussion is ridiculous.
Look to the Fed-Treasury accord, the fed set a ceiling on long-term interest rates of 2.3%. That is all you need to know. http://www.richmondfed.org/publications/research/economic_quarterly/2001/winter/pdf/hetzel.pdf.
Posted by: BD at September 14, 2010 01:15 PM
BD - They may be perfect substitutes or not (aren't T-bills sold at a discount?) but it is not all that is done.
Isn't the Fed supposed to work for FULL EMPLOYMENT also?
Posted by: Adam at September 14, 2010 01:31 PM
This has been fascinating. A reader could spend a whole day dissecting all these comments. I found the prof's comparison of Fed QE to the Treasury simply stopping issuance of long bonds to be very interesting.
My minor contribution is this: all the above assumes the only thing the FRBNY can buy is T-notes. This assumption is made because the large majority of new mortgages are backed by Fannie/Freddie, therefore resumption of FRBNY purchases of CMOs wouldn't really help.
Well, purchases of T-notes only helps the govt. continue ridiculous deficits. Some have mentioned Japan. During Japan's QE, the amount of JGBs held by the banks _increased_. They simply took the QE proceeds and bought new JGBs. I suggest the FOMC instead pursue purchases of debt that is neither govt nor mortgage.
Also, the US has the world's most robust corporate bond market. E.g.- Goldman Sachs issued bonds paying 1%. I conclude that the govt., the mortgage issuers and the (non junk rated) corporations already have ready access to huge amounts of super cheap financing.
The FOMC should consider buying deliberatly lower rated long term instruments in order for QE2 to have a meaningful effect.
Posted by: JJ at September 14, 2010 11:56 PM
Wow. Whatever happened to letting the markets work based on individual preferences rather than using bureaucratic actions to force activities that are not supported by the fundamentals?
Is the Fed supposed to create another bubble that we will have to recover from some times in the future? The Fed's meddling created the problem in the first place so let us not make things worse by encouraging it to do even more of what it has been doing.
What is needed is a discussion about ending the Fed and letting the market deal with the supply of money and credit just as it deals with the supply of food, clothing, automobiles, etc. The government should stay out of the way and let individuals make voluntary transactions based on their own personal goals.
Posted by: Vangel at September 15, 2010 07:37 AM
We can't and shouldn't "end the Fed". The US being the only country in the world without a central bank? Also check the history of banking and the economy prior to 1913. It's quite scary.
What we need is a Fed that acted more like they did in the '80s, 90s and maybe '50s. That is to say stop managing markets, the world economy, fiscal policy and presidential and congressional elections.
Granted what they are doing now has been deemed a necessary "response to crisis", but we need more focus on how the "crisis" got here.
Unfortunately, Basel 3 and financial reform is looking like they are only kicking the can down the road for us to trip over in the not so distant future.
Posted by: Cedric Regula at September 15, 2010 11:27 AM
colonelmoore: "Don you said that you agreed that demographics would require Japan to cash in some of its foreign investments. You can think that but you can't agree with me because I never said that. I said that the government (not Japan which is composed of private entities as well) will have decreased tax revenues, and so it will have to finance its obligations with more borrowing. Because people are becoming net dissavers, this will have to come from foreign buyers of JGBs. And they won't want to lend for the same low interest rates in such a situation. Then Japan's government will be in a debt spiral."
Well, I admit to playing fast and loose with the difference between government and private sector in Japan. So let's reconsider. First, take the line "Because people are becoming net dissavers, this will have to come from foreign buyers of JGBs." Note, however, that Japanese citizens can move wealth from one asset to another, they don't need to be net savers to buy more JGBs, so the new JGB purchases don't have to be bought by foreign lenders. That won't negate your point that interest rates on JGBs will need to climb to induce this transfer. However, such transfers, and people drawing down foreign assets for consumption (my demographics point), will cause yen appreciation. The net result could be that the interest rate on JGBs may fall in future, despite lower tax revenues and higher government borrowing. That was my point. You can't look at just one result of the demographic changes and conclude from that what will happen to borrowing costs on JGBs. Furthermore, the higher yen values may be in keeping with full employment of a smaller work force.
Of course, Japan could simply cash in some of its massive holdings of foreign reserves, but my guess is that they will raise their consumption tax rate instead. That is, I doubt this: "These things take time but the collapse will absolutely happen." Since the debt is entirely internal, the only bar to solvency is difficulties in internal redistribution. You are assuming implicitly that inflation is the only viable mechanism to accomplish this. In the case of Japan, I disagree.
Posted by: don at September 16, 2010 02:51 PM
Deficit monetization would obviously lead to a new, much worse financial crisis than 2008, but we've been over that. What I'm surprised by is the complete lack of mention of the risks to the private sector of low long-term interest rates.
Jim, if you're interested in the issue of policy vis-a-vis long-term rates, you must come to grips with the concept of bond duration (http://en.wikipedia.org/wiki/Bond_duration).
When rates rise, low-interest, long-term debt falls dramatically in price. The more and the longer that long-term rates are suppressed by policy, the more long-term, low-interest debt is accumulated in the economy, and the larger the scale of the losses on those debts when rates rise. This is already a serious threat to household retirement savings and the solvency of the financial sector.
Posted by: Tom at September 18, 2010 06:04 PM
Didn't notice your comment until this thread got quite cool, but that's my biggest worry for the long term. Let alone what it means for personal investment, 401Ks and pension plans. But nearly all credit is securitized now, so this is like a bunch of hand grenades waiting to go off on financial sector balance sheets if inflation/interest rates go up for any reason, good (economy gets better) or bad(credit risk increases).
Posted by: Cedric Regula at September 20, 2010 12:13 PM
I'm with you. Awesome job! MMT explains so much. I wish would people would think before they type in response. Professor Hamilton's response about taxes was a classic!
Posted by: Detroit Dan at September 21, 2010 01:32 PM