————————————————————————————————

Monthly Discussion

 

 

The American Economy Revisited

One Year Later

 

The Newsletter issue of October 14, 2002 discussed (and forecasted) the American economy in terms of three parameters: GDP growth, unemployment, and interest rates. One year later we are revisiting the subject to see how the forecasts fared and to make new ones.

As usually, my approach is that the American economy is growing like an organism, thus following natural-growth curves (S-shaped patterns). The three indicators considered below, are among the most frequently used to track the economy. In my effort to cover as long a timeframe as possible—meaning the longest stretch of time amenable to an S-curve description—I use data all the way back to the end of World War II. Decidedly, WWII was a singular event that set new natural-growth trends.

Exhibit 3 shows the post-war growth of GDP in America. The S-curve tacks the data points remarkably well and is completed to the 68% level. Consequently the rate of growth, depicted in the lower part of the graph, is declining. The last four quarters, shown by the little circles, confirm the trend and the forecasts of last year.

 
                                    Table 1
                 Forecasts                    Actuals

                 

Year.qt

GDP

% growth

 

 

 

 

2002.q3

10382.84

0.9%

10506.2

1.2%

 

 

2002.q4

10472.07

0.9%

10588.8

0.8%

 

 

2003.q1

10560.55

0.8%

10688.4

0.9%

 

 

2003.q2

10648.26

0.8%

10793.9

1.0%

 

 

2003.q3

10735.17

0.8%

 

 

 

 

2003.q4

10821.27

0.8%

 

 

 

 

2004.q1

10906.55

0.8%

 

 

 

 

2004.q2

10990.98

0.8%

 

 

 

 

2004.q3

11074.55

0.8%

 

 

 

 

2004.q4

11157.26

0.7%

 

 

 

 

 

In view of the confirmation, no update of last year’s forecasts is necessary.

Exhibit 3.  Quarterly data. The small circles show what happened since last year. The purple line is an S-curve fit to the data. The goodness of the fit in the top graph argues for the validity of the forecast. The lower graph (deduced from the top) shows the life cycle of the process.

 

          The second indicator is unemployment shown in Exhibit 4 in terms of the Employment-Population ratio.

Exhibit 4.  Annual data. The evolution of US employment seems to be cyclical around the

S-curve that describes the post-WWII growth phase. The white circle represents year 2002.

 

          Employment in the US (in other countries too) is cyclical and rather regularly so. Year 2002 came below the trend, as had been forecasted, even if a little lower than forecasted. Exhibit 5 extracts and forecasts this cyclical variation. The 2002 data point came a little below last year’s forecast and more in agreement of downward excursions of previous years.         

Exhibit 5.  The green line is the black one displaced by one cycle. This way we obtain the forecast (red line underneath the green one). The little circle is year 2002, in line with downward excursions of previous years.

 

Table 2
                 Forecasts (last year)         Actuals             New Forecasts

Year

Employment to population ratio

Annual % change

 

 

 

 

2002

63.37983

-0.6%

62.4

-2.2%

 

 

2003

63.40474

0.04%

 

 

62.43

0.05%

2004

64.05384

1.02%

 

 

63.08

1.04%

2005

64.30176

0.39%

 

 

63.34

0.41%

2006

64.45865

0.24%

 

 

63.52

0.28%

2007

64.93283

0.74%

 

 

64.01

0.77%

2008

65.11312

0.28%

 

 

64.21

0.31%

2009

65.20382

0.14%

 

 

64.32

0.18%

2010

65.30616

0.16%

 

 

64.45

0.19%

 

The new forecasts are only marginally different than last year’s. Unquestionably employment is presently on the rise and will continue so at least until 2010.

The last indicator is interest rates. Exhibit 6 shows the data since the early 1950s fitted to a bell-shape curve. The reason for a bell-shaped curve is that interest rates reflect the rate of growth (investments themselves may grow along S-curves). Once again, we find that last year’s forecast was rather accurate (the actual data came a little lower) and the growth process is practically completed (96%), meaning that the interest rates should not change significantly from the 1.5 level for a long time to come. Table 3 compares last year’s forecasts with actuals.

Exhibit 6.  Monthly data. The purple line is a bell-shaped curve fit to the data. The little circles indicate the last 12 months.

 

Table 3

 

Forecasts (last year)

Year

Interest rate

 

Actuals

 

Oct-02

1.83

 

1.75

 

Nov-02

1.81

 

1.34

 

Dec-02

1.79

 

1.24

 

Jan-03

1.78

 

1.24

 

Feb-03

1.76

 

1.26

 

Mar-03

1.74

 

1.25

 

Apr-03

1.72

 

1.26

 

May-03

1.70

 

1.26

 

Jun-03

1.69

 

1.22

 

Jul-03

1.67

 

1.01

 

Aug-03

1.65

 

1.03

 

Sep-03

1.63

 

 

 

Oct-03

1.62

 

 

 

Nov-03

1.60

 

 

 

Dec-03

1.58

 

 

 

 

 

CONCLUSIONS

          Among wide speculations that American economy may be on the rebound, the above evidence corroborates indeed that employment is on the rise and will continue so until 2010. Also that interest rates will grow a little as they are presently below the level of 1.6, where they should be according to the natural curve. However, the rate of growth of the GDP is expected to go slowly but steadily down, and continue doing so … indefinitely! The top graph in Exhibit 3 is such a textbook like S-curve, that in no way justifies expecting deviations from the projected course. The only way to envisage important deviations is following unnatural circumstances. “Unnatural” in this context means circumstances unlike any other encountered during the historical window. The growth wave we are presently riding was triggered by WWII. So only events comparable in importance to WWII can launch the economy on a new major S-curve. The Iraq war and 911 may not qualify. In the absence of the “missing” war, stock market frequenters will have to rely upon short-term market excursions upward or downward, which will always be there.