January 7, 2004
At this time of year, I traditionally review my outlook
from the previous year to determine how well I understood economic conditions
and to see if I can learn something from my forecast errors. Under normal circumstances, my forecast performance was
worthy of a high B.
But there was nothing normal at the beginning of last year.
Venezuela oil production had collapsed from strikes that were to go on
for another two months. North Korea
was threatening to use nuclear power to force the U.S. to the bargaining table. And we were turning up the heat on Saddam Hussein in Iraq.
Under such circumstances, I believe it is only fair to give
me a low A for my forecasting efforts.
In doing this analysis, I am using comments from my
mid-January outlook column as well as the monthly forecasts I post on a
At the beginning of the year, I was confronting the reality
of lost momentum in economic growth, a backslide in employment (we lost another
800,000 jobs before new hiring finally returned in late summer), and further
uncertainty in tax policy.
I expected jobs and corporate spending to begin growing by
summer, and they did. While I
expected economic growth to be sluggish in the first half of the year, about 2
percent, I saw growth reaching 4 percent before year end.
I knew that Federal Reserve activity and tax policy were
unusually stimulative. For this
recovery to collapse with so much policy thrust would almost deny my
understanding of macroeconomics. Indeed,
my forecast was lower than what I thought that thrust could produce.
My failure to fully accept my economic understanding was the greatest
failure of my forecast.
In fact, the 2 percent first half growth was only slightly
too conservative. Of course, we
zoomed past 4 percent in the summer and should stay above 4 percent this fall.
Because of Venezuela's oil problems, I saw inflation
surging in the winter but then subsiding during the summer to less than 2
percent. In fact, prices did surge
and then fall. In November,
consumer prices were only 1.8 percent above previous year levels.
I suggested that the Federal Reserve might even "lower
rates" in the winter. They did
so in the spring and then held 1 percent for federal funds rates for the
remainder of the year.
I expected ten year government bonds to yield about 4
percent all year. In fact,
deflationary concerns pushed rates down to nearly 3 percent in June.
By the end of the year, however, yields are slightly over 4.1 percent.
I think I should be excused for not seeing the deflationary
flash before reasoned thinking restored more appropriate yields to long term
rates. I mentioned that mortgages
would continue to fall, which certainly occurred until summer.
I took a lot of criticism for suggesting that the stock
market already was in a rising channel as the year began.
In fact, the October lows were tested but held in March, and then the
bull market surged in earnest. By
the end of the year I was looking for a "sustained rally to challenge
10,000….late in the fall."
As far as stock market forecasting is concerned, that is a
direct hit. I hope my readers saw
as much improvement in their investments as I did this past year.
Wages grew less
than I expected, but my use of the word "anemic" for wage increases
Unemployment only reached 6.3 percent before subsiding (I
expected 6.5 percent), but I again captured the essence of labor market
Consumers spent more robustly than I expected and the
savings rate was lower. I worried
about auto markets being satiated. That
worry materialized in the fall, but not as dramatically as I thought.
While I certainly did not paint a perfect picture of the
2003 economy, I certainly had all the correct outlines, and some of the detail
was remarkably close.
So what did I learn? First,
even if corporations tell you they do not see improvement, use your training and
trust economic theory. Theory
certainly did not fail me this time.
Second, unless "geopolitical" issues directly
impact economic conditions such as oil supplies or trade deficits, their impact
upon the economy is fleeting. Economic
behavior is more robust than most people want to believe.
And I still am a pretty good forecaster. So let me enjoy my low A, even if there were a few rough spots when you look more closely at the 2003 forecasting portrait.