Archive for January 2009
As the title says, this one is just for fun: a chart of the price of CDS on 10 year US Treasuries from April 2008 through January 27. The CDS priced at 68 bps today, after having hit a record high of 75 bps in the first week of December. According to FT Alphaville, the same contract trades around 2 (!) bps points in normal times. I should mention that the concept of CDS on US sovereign debt is somewhat flawed in my opinion (Who buys these things? Do people really think their counterparty will be around to honor the CDS if the US defaults? Isn’t the market too thinly traded for prices to be meaningful?), but their development is still interesting to watch. The chart below includes annotations on a few major Wall Street and Main Street events.
If you’re interested in commentary, I recommend the recent pieces by Greg Ip, Dean Baker and Felix Salmon for further reading. And if you have any thoughts on how sensible or flawed the notion of CDS on US Treasuries is, post a comment and let us know!
The government just successfully brought $40 billion of 2 Year Notes to the market, in the largest such auction in history – with the highest bid/cover ratio (2.69) since fall 2007 and a yield of 92.5 bps or 1.4 bps lower than anticipated by traders surveyed by Bloomberg, but markedly above the 88-89 bps range the when-issued notes where trading in. In reaction to the strong bid/cover ratio, Treasury prices are rising again.
So people are still buying Treasuries. That’s good news for the US, but I’m still worried about current price and yield levels especially given the increasing supply of bonds. To provide visual context to some previous posts about the Treasury bubble question, I am posting historical yield charts below. Take note of the previous record low yields (during the period of 1977 through the end of 2007) and the recent historical average yields since January 1990, which exclude the years of extraordinarily high interest rates in the early 80s.
90 day T-Bills: currently in the 10-20 bps range after having dipped into negative territory in December
2 Year Treasury Notes: trading below 1% while the recent historical average is 4.7% (since Jan 1990)
10 Year Notes: more than 300 bps below recent average levels (1990-2009)
30 Year Treasury Bonds: prior to 2008, the long bond had not dipped below 4% in decades
While I have come across a number of solid arguments for why Treasuries may actually be priced close to their fair value today, what comes to mind when I look at the above charts is “mean reversion.” What part of the puzzle am I missing, or what piece of data am I not paying enough attention to?
As I’ve been asking colleagues and friends what topics they’d like me to address here, I got a number of interesting nominations: from a closer look at the Iceland debacle over economic damage control to the question of what needs to happen for things to finally get better. I plan on spending time on all of the above in the very near term, but the one topic that currently occupies me is closely related to the previous posts – at least in substance, if not geographically.
I believe that inflationary tendencies are a significant concern for the US over the medium term. The same holds true for Europe, but the stakes are different. Not only does Europe have to worry about its economic future, but the economic union also has its common currency to look after. Starting last fall, talk of the EU’s Economic and Monetary Union (EMU) potentially falling victim to the financial crisis began to surface as European spread levels came to indicate that the currency union should not be taken for granted. In addition to widening spread levels, prediction markets are trading futures on the likelihood of at least on Eurozone member leaving the currency union by the end of 2010, and the prices indicate a likelihood of more than 20% of such an event occurring, with prior highs of more than 30%.
To take a step back from prediction markets or bond market realities (i.e. spread levels) and understand the macro mechanism behind what we’re seeing, I once again revert to textbook simplicity – what are the advantages and disadvantages of a currency union?
Primary advantages of a currency union per Harvard economists Robert J. Barro and Alberto Alesino:
(1) diminished transaction costs for goods & financial services (scalable with the number of member countries)
(2) peripheral members “import” the inflation rate of the anchor country (or anchor countries, in this case Germany and France which account for 45% of the Eurozone’s population and 52% or $4.4 trillion of the Eurozone’s $8.4 trillion combined GDP in 2006)
Primary disadvantages of a currency union:
(1) loss of flexibility in monetary policy, especially for peripheral members
(2) loss of seigniorage revenue unless elaborate currency union treaties are in place
(3) import of inflation rate for peripheral nations does not necessarily translate to price stability
What makes the Euro vulnerable to the current crisis is that it is an EMU product only by name – while the fifteen nations of the Eurozone are bound by a common currency and interest rates are set by the the European Central Bank, the union is a monetary union only and not an economic union. The impact of and responses to the global financial crisis vary from country to country, and quite drastically – yet all Eurozone members are subject to the ECB’s interest rate policy. As inflation expectations and the strength of individual countries’ banking systems diverge, we may come to a point where the advantages of sovereign monetary policy may outweigh the disadvantage of increased transaction costs.
William Buiter recently articulated a different view in the Financial Times, as picked up by Yves Smith at Naked Capitalism on January 15. Buiter suggests that “a eurozone country defaulting and leaving the euro is close to an unthinkable event.” What do you think?
I’m deliberately being simplistic and reducing the inflation/deflation debate to the monetarist framework; I think it’s useful as a starting point in order to get a grasp of the facts – and I like scary graphs. So take a look at the St. Louis Fed statistics regarding currency in circulation and monetary base: despite abundant talk of the Fed printing money, currency in circulation on an absolute level has not increased that drastically over the last few months. Sure, the slope of the curve or pace of increase has accelerated a lot, but from an absolute perspective things don’t look too bad:
Unfortunately, the above graph only shows currency in circulation, or the amount of currency available to consumers. The picture looks very different when you look at the US monetary base (i.e. Fed reserves and the reserves of commercial banks at the Fed): the monetary base has more than doubled over the course of 2008.
Hard to argue that this looks like it should spell inflation. Why haven’t we seen any of it so far? The answer is simple: banks have increased their excess reserves and are too scared of what may come to lend to the consumer. The graph below shows the missing link between monetary base expansion and the growth (or lack thereof) in the volume of currency in circulation, namely bank reserves in excess of Fed requirements.
The above charts suggest that we are currently seeing deflation concerns because banks are increasing their excess reserves faster than the Fed is expanding the monetary base; in other words, all the additional money printed by the Fed is getting soaked up by banks and the consumer is seeing decelerating inflation. This trend is bound to reverse suddenly and drastically once credit thaws, though. At that point do you think the Fed will be able to drain excess cash from the system fast enough to prevent massive inflation?
The Fed has pledged to step in and buy Treasuries if necessary in order to keep long-term yields down to make credit affordable, resuscitate the housing market and support consumer spending. But as issuance is increasing and foreign demand for Treasuries decreasing, will the Fed be in the position to purchase Treasuries in a sufficient volume to keep rates down?
- The 2009 deficit is expected to top $1.2 trillion – almost three times larger than the previous record at $440b in fiscal 2008.
- China, which accounts for ~25% of foreign (and >5% of overall) Treasury ownership, shows decreasing interest in US government debt: its foreign reserves were reported as shrinking in December ’08 and are expected to increase by only $177b this year compared to $415b last year.
Is the Fed going to be able to keep yields at their compressed levels for an extended period of time? I’m not convinced.
There has been much hype recently about the alleged bubble in US Treasuries. After returning more than 20% in 2008, Treasuries are now commonly viewed as a risky investment. Merrill Lynch and Goldman Sachs are among the few who believe that Treasuries are not overly expensive but fairly valued. More or less the rest of the world believes that the Treasury bubble is about to burst. And there is plenty of ammunition for the Treasury bubble view:
- Treasury yields are at historical lows
- issuance is increasing drastically as various domestic bailout packages have to get financed
- money supply in the US is increasing at an unprecedented speed, suggesting the dollar may weaken due to simple over-supply
- a weakening dollar, in combination with extraordinarily low yields, will decrease foreign demand for Treasuries
- Chinese demand for Treasuries will drop as China is struggling to finance its own domestic bailouts
- domestic risk appetite may increase at any time, leading to an investor exodus from the Treasury market.
I am sure that, over time, demand for Treasuries will weaken and yields will mean revert resulting in a sharp correction in prices. Hence my general advice to anyone holding Treasuries in their portfolio (i.e. hopefully everyone) would be to keep the exposure but shift the duration towards the very short end of the curve. I believe that Treasuries have run up in a huge way, and I do believe that they will come crashing down at some point – but I don’t know if it’s going to be tomorrow, in a month or in a year. In addition to not knowing when Treasury prices will correct, I like having Treasuries as a disaster insurance. I just don’t like the idea of potentially losing 20%+ on a long bond portfolio as yields are decompressing.
And here’s why I’m still a believer in Treasuries, even though I think the bubble is going to pop sooner or later:
There is a good chance that we haven’t felt the full effect of the credit crisis yet; if things get worse, there will be increased demand for Treasuries going forward. And while it is true that a weak dollar would decrease foreign demand, central banks around the globe are likely to be cutting rates over the next months as a response to the global slowdown (i.e. the dollar may remain comparatively strong). In addition – and this argument is lifted from Kessler, who recently made a smart case for Treasuries – demand for Treasuries may well increase once Obama puts together a concrete plan of action because political fighting will break out which will undermine confidence in a quick recovery and provoke a renewed flight to safety as the general outlook deteriorates. In addition, the US government has a number of incentives to keep rates as low as possible via the Fed funds target rate and quantitative easing:
- if yields in longer-dated Treasuries increase the housing market will continue to spiral downward as rates on fixed rate mortgages will increase (and so will rates on all other collateralized loans); this creates a negative wealth effect and translates into lower consumer confidence & consumer spending
- higher yields would have a doubly negative effect because borrowing is particularly important in times of high unemployment: credit needs to bridge the gap.
What do you think? Do you still hold Treasuries in your portfolio? And how quickly or slowly do you believe the US economy is going to recover?
I’m currently doing some research on the (likely) Treasury bubble, and the charts I’m coming across are nothing but scary – one more so than the other. Check out the following which shows the $$$ amount borrowed by US banks from the Fed through Dec 2007; the spike marks the Savings & Loan Crisis at the end of the 1980s with borrowing maxing out at $8b.
Now take a look at the following chart. It is the same graph as above, but updated through the beginning of November ’08.
This might be less scary if the Fed wasn’t creating money out of thin air and at the same time accepting assets of questionable – and deliberately undisclosed – quality as collateral from banks. As it stands, I am not surprised that 10 year CDS on US Treasuries are above 60bps (high of 72bps at the beginning of December vs. a pre-08 historical average of 2bps).
[The charts above are publicly available from the St Louis Fed; I first came across the idea of presenting the two graphs in progression in a letter by DK Matai.]