Jewish World Review March 6, 2003/ 2 Adar II, 5763
David R. Kotok
What will the larger deficits do to interest rates?
http://www.NewsAndOpinion.com | Many things influence interest rates. Forces ranging from Fed policy, the inflation outlook or currency weakness (strength) compete with business conditions, loan demand, savings rates, and even the stock market direction---all are constantly pulling rates up or down.
At any moment the market reflects the outcome of this dynamic tension. All rates do not move in the same direction and by the same magnitude at the same time. Rates differ by type of debt and credit quality and by country and in different currencies. A matrix of all rates by all maturities and all debt classes has thousands of entries. Each of them is changing in real time and each of them is always reflecting a consensus market-driven pricing. The spreads (ratios and relationships) among and between them constantly change.
For this commentary, we'll focus only on the benchmark 10-year U.S. treasury note and only on the U.S. federal deficit. That immediately means we are making the wrong assumption about everything else (hundreds of influences). We are assuming that all the other forces remain unchanged. We do that knowing that in the real world none of them are static. The interest rate beast is an extremely dynamic animal.
Estimates of the impact of the federal deficit (or surplus) have a wide variance. Perhaps the smallest is from the Bush Administration. Their economic work suggests that every $100 billion of additional federal deficit will raise the benchmark 10-year note interest rate by 1 1/2 basis points (a basis point is 1/100th of 1%). Add 300 billion to this year's deficit and you add about 5 basis points to what the 10-year note yield would otherwise be.
This Bush Administration estimate seems too low to me.
At the other end of the spectrum there are some folks who believe that a couple of hundred billion of additional deficit will raise the benchmark 10-year rate by a full percentage point. Other folks convert nominal rates into real ones (by removing a measure of inflation); they then estimate the impact of the deficit on real rates after which they add back their own inflation assumption to get an estimate for the nominal rate.
There are many estimates using variations of these approaches. By the most extreme methods, a 300 billion deficit will add nearly 200 basis points to what the 10-year yield might otherwise be.
We think that this is too high an attribution of the impact of the deficit.
In our shop we start with the present nominal interest rate. For today, let's call the benchmark 10-year treasury note at a 3.75% yield. In our work we have found that changes in the market value of the treasury note over time act as an influence and help determine the other boundaries or limits of interest rate movement. To measure those changes and boundaries we have to adjust for inflation. After complex adjustments which remove the inflation component of the interest rate, we derived a real holding return. We get that by measuring the real interest rate and the real price change and combining them.
Our work is distinguished from the others because it incorporates the price change of the 10-year note into the computation. We do not just look at the yield. This combination leads us to the cyclical limits of interest rate movements in real terms. We can now measure them and forecast them with some reliability and historical support.
We continuously look at where real interest rates are and how they have moved in the cycle. That helps us determine a forecast for real rates which, in turn, leads us back to nominal rates. This method also guides us in positioning our bond portfolios because it gives us our direction and target maturities on the yield curve.
We still have not added the deficit impact. We'll get to that soon but first our estimate without the deficit impact.
Today, out method suggests that the real holding return on the 10-year benchmark treasury note has reached a level which is extreme. A half century of history suggests that this is a time (and level) to sell it and to shorten maturities and become defensive in strategy. Using this method we can project where the interest rate must go during the next year or so to complete a typical movement. Our method tells us that the yield on the 10-year treasury note should tend toward 5%-to-5 1/2% from its current 3.75% level.
Bond prices move inversely to interest rates. At the present level, an abrupt 100 basis point rise in the market rate would take nearly 9 percent of the capital value out of an existing 10-year treasury note. In other words, at today's low interest rate it would take nearly 1/4 the life of the 10-year note to recover the capital loss of an interest rate movement which could occur quickly. That is a very high risk profile in our book.
History shows us that ignoring this warning can be perilous. Investors who bought bonds during times like these and at valuations like this subsequently watched those bond prices decline as interest rates rose. Sometimes these movements were violent; in other times they were mild. At all times it would have been better for the investor to avoid the bond purchase and wait.
We have tested this forecast level of 5%-to-5 1/2% with another method. We estimate a normalized growth rate for U.S. Gross Domestic Product (GDP). We also estimate an inflation rate using a variety of inputs. Today, we are guessing that the U.S. economy will grow at 3% and that inflation will average about 2%. We get our interest rate forecast by adding the two items (growth plus inflation) together.
This exercise tells us that the targeted yield on the 10-year note today is around 5%.
But what about the deficit?
Our work reviewed the entire post World War II period of surpluses and deficits. From what we can glean, we think that a change in the deficit (surplus) can raise (lower) the 10-year treasury real yield by between 5 and 15 basis points per 1% of GDP. We cannot do much better than that because of the other influences we mentioned at the beginning of this commentary. With Iraq in the mix, we expect the deficit impact for the next year will put the 10-year yield 20 to 60 basis points higher than it would otherwise be.
All this leads us to the following conclusion. The current 3.75% yield is much too low if the economy recovers and the stimulus in the pipeline works. If it does work, we expect the 10-year note to eventually yield above 5 1/2% and perhaps as high as 6% when a 300 billion plus deficit is fully factored into the market.
What if we are wrong? What if the market yield of 3.75% is accurately forecasting the future?
If the market is pricing the 10-year note yield accurately now, it is suggesting that the economy will not recover. Furthermore, a current 3.75% 10-year treasury note yield implies that economic growth will be at a hesitant recession level of under 2% for a while and that inflation will remain under 2% and trend lower toward 11/2%. While this is possible, we do not consider it to be a likely outcome now that the Fed has committed itself to avoiding such a prolonged period of economic weakness.
The present 3.75% treasury note yield is a market forecast that the three prongs of stimulus will fail. We have a very low Fed Funds Rate, a growing fiscal stimulus through the deficit and tax cuts---the three combined are enormous stimulus. The combination has not been seen in anyone's living memory. If this fails, it will be a first in America economy history.
We believe that betting against the Fed and the other stimulus is a losing proposition. We won't do it with client's funds; we wouldn't do it with our own money.
This leads us to our very cautious strategy toward bonds. Both our taxable and our tax-free bond portfolios are
very defensive. We expect a dramatic rise in interest rates over the next year or so as the stimulus works and
as the deficits start to materialize.
02/24/03: There are 2 Two Front Wars!