June 01, 2009
High Anxiety (about Interest and Inflation Rates)
In March 2001, I was tasked to follow developments in Japanese macro policy (including monetary, exchange rate, and banking recapitalization issues). Readers will be tempted to ask what this has to do with current events. Well, at the time, Japan was facing rapidly rising net debt-to-GDP ratios (rising from 60.4 ppts of GDP to 84.6 ppts from 2000 to 2005), and was embarking upon a policy of quantitative easing in an attempt to stave off a deep recession. And yet opponents of quantitative easing worried about hyper-inflation, even as y/y inflation at the time remained mired in the negative range. I didn't understand the fears at the time; and I still don't. Now flash forward eight years, and move across the Pacific.
Recent commentary has focused on the sharp reversal in US long term interest rates  and the steepening of the yield curve  . The anxiety seems to be centered on the collision between increasing supply and decreasing demand for US government debt (although see Brad Setser's recent article), combined with quantitative easing that spurs fears of debt monetization. US Federal debt held by the public to GDP ratios are projected to rise from 56.8 ppts to 73.2 ppts (end-FY09 to end-FY14) (Source: CBO, March 2009), while reserves have gone from $96.5 billion in August 2008 to $949.6 billion by April 2009 (seasonally adjusted figures, from FREDII). Of course, returning to my Japan case, reserves (actually, the "current account balance") rose from about 5 trillion yen to about 32 trillion by 2004 during the episode quantitative easing, before reverting to about 9 trillion in 2006 (Source: Humpage and Schenk, 2009). And yet, Japanese y/y inflation never really stayed in positive territory until 2008 (and it's back to negative range in 2009Q1).
Hence, I thought it useful at this juncture to review trends in long term nominal interest rates and long term real interest rates, in order to get some perspective.
Figure 1: Five year (blue) and ten year (red) constant maturity nominal yields, monthly averages of daily data. Triangles denote data for May 27. NBER defined recession dates shaded gray. Dashed line indicates sample start for Figure 2. Source: FREDII and NBER.
Figure 2: Five year (blue) and ten year (red) TIPS constant maturity yields, monthly averages of daily data. Inverted triangles denote data for May 27. NBER defined recession dates shaded gray. Dashed line indicates sample start for Figure 2. Source: FREDII and NBER.
Ten year constant maturity rates as of 27 May are back to levels of September 2008, as are the real rates (keeping in mind the distortions in the TIPS markets). The sharp rise in the nominal rates, even while real rates are only slightly above where they were in April, suggests to me that about half the nominal interest rate movements are due to revisions to inflation expectations (here's Jim's take on the inflation component to these yield shifts).
Taking the Treasury-TIPS spread literally as a measure of inflationary expectation (see caveats in the appendix here), 10 year expected inflation was 1.36% in April; as of May 27, the implied annual inflation was 1.88%, an increase of about 0.5 percentage points. The increase in the 10 year real rate was 0.26 percentage points. Of course, one would want to allow for a lot of uncertainty in these calculations, given the academic research detailing the problems with equating break-even with expected inflation (see D'Amico et al. 2008), even before the recent turmoil in the TIPS markets.
Of course, the United States in 2009 is different than Japan in 2001. One key difference is that Japan was, and remains, a net creditor. America is a big net debtor to the rest of the world, with extremely large holdings of US Treasurys by foreign private and state actors. And so, for me, I worry more about higher real interest rates (portfolio balance effects) than higher inflation. But even here, real yields according to TIPS seems fairly low in historical perspective (and roughly comparable to those prevailing during the period characterized as "the saving glut").
My bottom line: Think, and recollect, before panicking.
Posted by Menzie Chinn at June 1, 2009 09:02 PMdigg this | reddit
To be honest I don't think Inflation will be all that devastating. Do I think it will go pretty high? Yes. But I do not see it getting over 5-6% at the very most. I do think Gold is a good investment in small quantities as protection.
Posted by: Trump at June 1, 2009 07:16 PM
I agree with you that higher real interest rates are a more tangible worry than consumer price inflation. However, with our elevated level of indebtedness, the Fed will face irresistable political pressure to counter increases in interest rates by continuing to create money. Inflation is the path of least resistance. The government has made promises it cannot keep. True to its dishonest nature, it will inflate these obligations away.
"real yields...seem fairly low" What else would you expect during the initiation of 'quantitative easing'?
Posted by: algernon at June 1, 2009 08:37 PM
I think hyperinflation is being devalued to 4-6%. What do they think will happen when the Fed starts selling those assets it has acquired, or do they think they plan on holding them permanently?
Posted by: Lord at June 1, 2009 11:25 PM
fine....if the US doesn't want a zombie economy....but I think they want one that's a little more robust than that.
Posted by: apj at June 1, 2009 11:44 PM
The economy is downsizing to a lower growth trend and the state and federal governments are late in the process.
Posted by: Mattyoung at June 2, 2009 04:52 AM
I believe high inflation is coming. The reason is the growth in M1. http://research.stlouisfed.org/fred2/graph/?s[id]=M1. M1 has gone from about 1400 to 1600 in about 9 months, a 14% increase. I dont believe Japan had such a large increase in M1 over so short a time period which is why they avoid inflation, but if I am wrong on this, please send me a link. Also, I believe it is more likely that the fed will buy even more treasuries as the budget deficits grow even larger. The fed will be required to print the money to buy these treasuries which will expand M1 even more. Who else will buy the 2 trillion in treasuries over the next year? Perhaps the Chinese, but with 750 billion in treasuries already, I dont think they are they likely buy much more. The possibility of hyperinflation arises if the rest of the world wishes to sell their treasuries and the fed is forced to purchase all of these plus the new treasuries in order to try to keep interest rates artificially low.
Posted by: Jeffrey Knoll at June 2, 2009 06:29 AM
Japanese households were net savers, whereas American households are net debtors - inflation is beneficial to net debtors, but harmful to net savers, so Japan didn't have the same politcal motive to create high inflation. Japan also waited until deflation was well entrenched, and was far less aggressive and creative than the Fed has been in terms of pushing money into the banking system. Except for the wild-eyed gold bug crowd, I don't think a serious economist or bond trader looks at a snapshot of today's situation and sees imminent inflation. However, looking forward, it is doubtful the US will be able to reduce the structural federal deficit to something more sustainable, like say 3% of gdp, or that the Fed will be able to unwind the massive monetary stimulus it is currently pushing into the system. Whether a meaningful pickup in inflation eventually occurs is an open question, but there are good reasons not to dismiss the idea. Besides, in finance, when the crowd decides to panic, it's never wrong to be the first one to hit the exit button.
Posted by: Turbo at June 2, 2009 06:43 AM
With all due respect, it sounds like you are saying, "this isn't going to happen because it hasn't happened yet."
What's important is the relationship between policy actions and interest rates. As long as deflation fears dominate, policy makers can administer large doses of stimulus and bank bail outs with zero impact on the government's borrowing cost.
However, when fears of structural deficit monetization rise, then each additional dose of stimulus comes at the cost of a higher Treasury yield. Presto, the stimulus free lunch is over.
We may be entering that second phase now. If the economy again weakens, then ironically, bond yields may rise as the market anticipates an additional flood of Treasury issuance.
This dynamic is not as far-fetched as it sounds. It happens in emerging markets all the time. The fact that yields are rising now, in the face of obvious short-term deflation, is a potential signal that we are entering this emerging markets dynamic.
Panic? No. Hedge against inflation and higher rates? Most certainly.
Posted by: David Pearson at June 2, 2009 06:45 AM
Real wages for the majority of the working class have been in decline since 1969. Women have made gains and minorities as well, but as a whole the working class has lost ground. This is one of the main reasons inflation is not a problem. In fact, the working class has hit the wall and was only maintaining their standard of living because of loose credit, which has now been taken from them. Even with gas, a big commodity for the working class, at $2.65 they are acting as if it were $4.25. Consumption is down and will continue to stay so until the debt load is relieved, either by paying down, defaulting or reorganizing.
Watch for a flat recovery even with the money being pumped out for the next election, ie. 2010.
When this money hits the street it will go straight back to the banks in the form of debt repayment and very little will actually work its way into circulation. 2011 and 2012 are not looking very good, right now, although with the 2012 election being the most important election since 1952, I am sure a third dumping of money will occur.
This is, currently a time, very similar to the 1930's in that the working class has had the hell scared out of it. Do not expect the babyboom or boomlet folks to ever allow this to happen again in their lifetime. It will be 20 years before this mess is cleaned up! Thus the 1952 election process.
Posted by: steve at June 2, 2009 08:15 AM
High real interest rates, declining value of the US dollar, slow economic growth, stagnant productivity growth, on-going loss of relative hegemonic power and the resulting privileges, bitter domestic rent-seeking competition. These are all possible.
I don't see where unusually high rates of inflation are going to come from. Where are the price-pass-along mechanisms? The concentration of market power that helped fuel inflation in the 1970s and 1980s is for the most part absent.
Posted by: GNP at June 2, 2009 08:15 AM
Posted by: ignorance arbitrage at June 2, 2009 08:32 AM
I'd be curious to see research on the effect of falling testosterone levels in the population affecting the economy. It's been suggestion that the financial bubble was fueled by attraction of relatively high testosterone individuals to finance and executive jobs. Active and risk seeking personalities.
But in the broader population testosterone levels have fallen. This is very significant because I believe it may have contributed to the decline in efficiency of our transportation and slowed the real economic growth. The financial bubble itself may not have resulted in crisis had it not been for the essentially non-existant economic growth during the same time (probably negative if you eliminate the direct effects of government spending and spending on financial instruments).
Not only does low testosterone decrease drive. It causes people to carry more weight. It cause heart problems (I won't be suprised to find out that some of the relationship heart disease risk we associate with is actually caused by low testosterone). It inhibits the ability to develope physical endurance and causes chronic pain, which compounds the problem of decreased drive and weight problems.
Posted by: aaron at June 2, 2009 10:39 AM
I think low testosterone is also related to high anxiety.
Posted by: aaron at June 2, 2009 11:08 AM
A site you reference argues that the overstatement in expected inflation gained from comparing TIPS and nominal bonds that results from the principle-protected nature of the TIPS is small, because deflation is highly unlikely. But Mankiw pointed out that this could explain a big increase in imputed expected inflation that coincided with a change in the mix of TIPS used to calculate the average return that included a much bigger share of new issues.
Posted by: don at June 2, 2009 11:32 AM
From my point of view, this is how you want QE to work against Debt-Deflation, which is a panic phenomenon.
1) Low Short Term Interest Rates, as a disincentive to buy guaranteed assets, and an incentive to buy stocks and corporate bonds. This attacks the Fear and Aversion to Risk.
2) Rising Longer Term Interest Rates, which signal an end to Deflationary Fears, and are an incentive for Longer Term Investing.As well, as you say, the Spread often ( which is as exact as this gets ) signals a recovery.
Now, what's interesting, is that I take it that this is what Bernanke and Geithner have been arguing, although not necessarily saying it the way I do. And, in fact, it seems to be working, which , again, is about as good as it gets, since none of us know the future.
Yet, as obvious as this is to me, others have been puzzled by this line of reasoning. I, on the other hand, do not understand the point of having interest rates move in tandem, insofar as incentives are concerned.
I understand the idea of rising interest rates being a problem, but the Fed has other means of addressing mortgages, in coordination with other parts of the government. However, I am not for keeping interest rates of mortgages artificially low at all, but certainly feel that this policy should now end. We need to see a plausible bottom on housing prices, without the perception of government still keeping housing prices artificially high.
Of course, Inflation will be the issue going forward, but I prefer that to a Debt-Deflationary Spiral. Call me silly, I guess.
Right now, a stimulus with QE is working as Geithner and Bernanke and I would like. It could all go sideways, but the fact that this is the plan seems to be missed by many people
Posted by: Don the libertarian Democrat at June 2, 2009 01:33 PM
The US and the world as a whole are currently going through a bout of deflation. Governments around the world are issuing a lot of debt to bail out insolvent banks, pass stimulus packages, and try to smooth out the recessions.
With every major industrialized country's economy dropping rapidly, which central government is going to step in to the breach and buy all this debt? As you note, the US in 2009 is not Japan in the 90s. In the 90s there was a general worldwide boom. How much debt can be issued by the world as a whole and by the US in particular before the appetite curdles?
Posted by: uber_snotling at June 2, 2009 01:54 PM
Apart from the historical experience of Japan, why wouldn't we expect that quantitative easing and the projected increase in government debt implies higher future inflation? One hypothesis is that Jim Hamilton and Gary Gorton have diagnosed the financial crisis correctly: it is a run on the shadow banking system. If this is so, then the effective (not official) money supply has dropped dramatically. Increases in the Fed balance sheet and projected increases in government debt are in some magnitude a replacement of this lost money supply, and simply keep the price level from falling.
Posted by: Robert Driskill at June 2, 2009 02:47 PM
You call a rate of 3.6 percent on 10-yr US government bond "panic"?
Let's see: the average 10yr CPI inflation has not been under 2% in the past 40 years, and the average for the 40yr period is 4.5%. The US is a net debtor country, as you correctly pointed out. Mainstream economists, such as Mankiw, are contemplating higher inflation through negative nominal interest rates, such as randomly destroying 10 percent of the currency per year ... panic? I'd say anyone is lending their money to the US government for 10 years at 3.6% is pretty brave, comparable only to these leading money to homebuyers at less than 5%, or banks at 0.2%, or ... well, a saver just has no place to put his hard earned money when the federal reserve is using all its imagination and power to keep interest rates low ... no wonder stock market is rallying, so are commodities prices ...
Posted by: Yu'er at June 2, 2009 02:54 PM
And again we see that reasoned and straightforward comparisons with some actual foundation in fact and example are immediately tossed aside in favor of convoluted doomspeaking scenarios. (Much as the fact that China is increasing its holding of dollars and debt means nothing in the face of angst about whether China has an appetite for dollars and debt.)
Posted by: PM at June 2, 2009 06:44 PM
Don the liberterian Democrat,
Your handle generates a certain degree of cognitive dissonance. Just what aspect of liberty is of any value to a Democrat?
Furthermore, the 'Debt-Deflation' liquidation (though not a 'Spriral')which you abhor is what would happen in a free society with rule of law, which is what I imagine a libertarian society to be.
Posted by: algernon at June 2, 2009 06:56 PM
I dont believe Japan had such a large increase in M1 over so short a time period which is why they avoid inflation, but if I am wrong on this, please send me a link.
That would be incorrect. Per data.un.org, Japan's M1 increased 22% in 2002 and 19.7% in 1997.
Printing money is not necessarily the end of the world, and at times like this, it can be a benefit if managed appropriately.
Most of the money in the economy is not in the form of M1, but debt. Deleveraging is as good as removing money supply from the system, and we've had considerable deleveraging. Given the lack of wage growth, it should be possible to print some quantity of money as an offset to the lack of credit without creating inflation. There is a limit to QE, but that limit isn't zero.
If growth and credit resume, that M1 growth can be curtailed or even reversed if necessary. If the powers that be are vigilant and don't attempt to reduce the debt by deliberately degrading the currency, it should be fine. It comes down to whether you trust the Treasury and the Fed.
Posted by: Angry MBA at June 2, 2009 09:36 PM
No, I do not trust the Treasury or the Fed. Nor any of the central banks that manipulated their currencies to remain export competitive. Nor any of the governments that ran structural deficits during boom years. Desciples of Keynes come rain or shine. Nor governments that trample on bond holders and discriminate against savers to curry favor with unions or other special interests during bankruptcy procedings. Nor any government or private company that runs persistant unfunded future liabilities with no plan to pay for those liabilities. The great moderation has ended. Now the real tough choices have to begin. Do I trust them to kick the can down the road? Yes, I do. But trusting the Treasury or the Fed to protect savers and ordinary taxpayers is a fool's bet. Caveat Emptor.
Posted by: MrBill, Eurasia at June 3, 2009 12:06 AM
Angry MBA: "If growth and credit resume, that M1 growth can be curtailed or even reversed if necessary."
Yes, but it depends, no? If there is a flight on the dollar, or if interest rates spike (and bond holders dump treasuries), the Fed will be selling (to mop up $'s) into a run, further doing what it does not want to do: drive up interest rates. Methinks Dr. H has posited on this in previous essays...
@Don the LD: "...but the Fed has other means of addressing mortgages, in coordination with other parts of the government..."
Ever heard of FNM and FRE? Don't you just love it when the market has to compete with an implicit (now explicit) guarantee by the govt (or with someone who can print $)
On topic and on thread two points:
1. My dad, who never finished high school is a buyer of gold. He doesn't understand economics other than what he sees as rising prices and diminshed purchasing power (in spite of his SS cola). Looking at the premium over spot to actually purchase bullion gives some indication of the amount of "panic" in the air by the masses no? Aren't expectations what matter?
2. The role of even "educated" people flows from #1. In a contract negotiation the other day with an ex-US company and who should bear the currency risk I heard, "you guys are printing money..." There are real business implications to QE, and they are not good for business.
Posted by: Jumbo Shrimp at June 3, 2009 04:44 AM
Angry MBA. Thank you for the link. I agree with you that if the fed reduces the money supply in the future, it will reduce inflation and it all depends on whether we can trust the federal government and the federal reserve to do this. However, I dont see this happening since we are facing a $2 trillion deficit this year, and about $1 trillion a year for the next 3 years. I dont think it is likely foreigners will buy this much of our debt, so I believe it will be left to the fed to monetize it.
Posted by: Jeffrey Knoll at June 3, 2009 05:41 AM
I do not trust the Treasury or the Fed.
I'm of the opinion that they've generally managed this fairly well. Their moves have generally been sensible, and their mistakes haven't been fatal. We should police them, of course, but I don't understand the propensity to assume the worst about them. It's not as if Mugabe is the Treasury secretary.
I dont think it is likely foreigners will buy this much of our debt, so I believe it will be left to the fed to monetize it.
That's a fair concern. (Damn, I guess I'm not that angry after all.) You may be right. I hope that you're not, but I would allow for the possibility that you are.
I haven't crunched all the numbers on this, but I presume that that the hope here is that low-yield guaranteed securities will be attractive enough to attract capital that may have previously sought higher or more risky returns during the go-go era that we just lost. In other words, it's a positive flip side to the concept of crowding out, except here, there is actually a demand for these securities, given a diminished appetite for riskier products.
Rates will probably track upward, but given the low starting point, the rate itself may end up being digestible. Without inflation, a 5-6% long bond might start looking pretty good to a lot of investors, and foreign governments still need a means to secure their reserves. On the whole, I'm inclined to agree with Mr. Chinn here, I'm just not prone to panic about this.
Posted by: Angry MBA at June 3, 2009 05:12 PM
Just to add some numbers, the total outstanding US debt to others (not internal like owed to social security) is about 5T. About half (!) is due in 2009. In addition there is about 2.5T of new debt to be issued (deficits etc). So the US treasury needs to sell about 5T of debt in 2009. Given that there are about 50 weeks in the year that's 100B/week for the whole year...
There isn't any room here for any but successful auctions...
Posted by: geoge smith at June 4, 2009 09:22 AM
Beyond just the academic debate over the inflationary impact of printing money, I think many economists except for some like Marc Faber are losing sight of the bigger political and economic trends.
At one time, I thought that nationalising the banks was the only solution. Now with the government taking over AIG, Citi and GM as well as the arrogant speech by Larry Summers to justify these actions, its quite obvious Obama's government is moving from unregulated capitalism to regulated socialism.
If this is the case and these nationalised companies are not auctioned off to private hands with a visible time frame, then the selldown of the US$ will accelerate because the Fed will need to print more money to bail out these black holes.
In other words, using debt to cure a debt-addicted economy is already a dubious remedy. What is worse is that the Wall Street oligarches are using this crisis to takeover the political system of the US before the public's eyes in the name of financial rescue.
Posted by: jeremiah liang at June 15, 2009 09:51 AM