Are We There Yet?
By Jamie Sumner
Ten minutes into our family vacation, which included a 15-hour car ride, the little voice of my four-year-old son asks “Are we there yet?” Although we attempted to prepare him for this long trek, he did not grasp what that really meant. After several hours of being in the car, and many more “Are we there yet” questions, we decided to approach his question by relating back to prior lengthy car rides, such as trips to the beach, amusement parks and grandparents.
This same question has been echoed throughout the banking industry about the shape of the yield curve. November was the sixth consecutive month in 2005 of a flat yield curve, and people are all the more anxious to know – are we there yet? Unfortunately, there is no correct answer because no one truly knows. However, what we can do is revisit history in an attempt to identify trends that might be relevant to our current interest rate environment.
TAKING A HISTORICAL PERSPECTIVE
This article examines the shape of the yield curve, between the two-year and 10-year maturities, to identify the time periods when the curve was steep, normal, flat or inverted. We will also present various factoids relating to each of these yield curve environments.
Please keep in mind that this article is not attempting to call interest rates and should not be used to identify where rates are going, for history is not always an accurate indication of future results. For this article, we have used the interest rates as of the last day of the month for the time period between June 1976 and October 2005 as our baseline because there are several distinct interest rate cycles during this time frame.
Before we attempt to identify these different yield-curve environments, we must first define a “normal yield curve.” To do this, we calculated the average positive spread between the two-year and 10-year constant maturity treasure (CMT) over our time frame, which was 102 basis points (bp). Then, to arrive at the normal range, we placed lower and upper boundaries equal to 50 percent and 150 percent of the average positive spread. A spread greater than the upper boundary we characterized as a steep curve, anything between zero and the lower boundary is considered a flat curve, and anything less than zero is considered an inverted yield curve. A summary of the definitions is provided in Figure 1 (below).
Once we defined the range for each yield curve shape, we were then ready to identify the different shapes throughout our near 30-year history. Figure 2 shows the total months, total periods (defined as one or more consecutive months of the yield curve shape) and the average length of each period. Thus, from Figure 2 (below), we can see that over this time period there has been a total of 87 months in which the yield curve was flat. Furthermore, these 87 months have been scattered across 14 periods with an average length of 6 months per period.
Because the current yield curve is flat, let’s carefully look at the 14 periods of a flat yield curve environment. Of these 14 periods, 7 lasted 3 months or less while the remaining periods lasted between 4 and 19 months (including our current flat environment which is 6 months so far).
Of the seven periods longer than three months, there were two instances where two of the periods were separated by a one-month inverted curve. If we were to count each of these occurrences as one period, we would arrive at four sustained flat environments over this time period (excluding our current environment). Figure 3 (next column) shows the lengths of these four periods and the shape of the yield curve succeeding the sustained flat periods.
OUR WORLD IS FLAT, FOR NOW
Figure 3 demonstrates that once a three-month period of a flat yield curve is breached, odds are we will be in that environment for the better part of a year or longer, based on history. Furthermore, we see that the succeeding curves have not always been inverted.
Notice that the longest period is the longest by a significant length (22 months). This 35-month period spanned between March 1997 and January 2000. During this time period there was a lot of economic activity which could have played a role. (For example, this was the period of the tech stock boom, gross federal debt as a percent of GDP declined after 15 years of expanding and the dollar amount of federal debt held by investors also declined.)
Looking at the three shorter periods, they average about one year in length. Also, two out of the three periods were followed by a normal curve. Thus, if history is a guide, we might pull out of this flat curve into a normal curve sometime in the second quarter of 2006.
WHAT’S AROUND THE CORNER
Currently, as of late January, the spread between the two-year and 10-year was just three basis points. The spread ended 2005 inverted by two basis points and the daily spread since then has ranged between zero and five basis points. So while we are not there yet, the end could be just around the corner.
Jamie Sumner joined BNK Advisory Group in 2000 as a financial analyst, helping to prepare BNK’s quarterly analyses, risk-versus-reward analyses, stock valuations and other analytical projects for both commercial banks and thrifts. In addition, he co-authored a paper on how banks can benefit from the implementation of Regulation D and is a regular contributor to BNK’s monthly newsletter and research pieces. Sumner previously held positions with Vista Bank N.A. He is a level III candidate in the CFA program and received a bachelor of science degree from Alvernia College, graduating cum laude. He can be reached via e-mail at firstname.lastname@example.org.