Are You Basing Your Decisions On Accurate Data?ByRaymond D. Matkowsky
The State Of Business
As a business owner or manager, you need some semblance of certainty as to what the state of business will be in six to twelve months down the road. The road is littered with businesses that were told everything is fine and there is enough strength in the economy to contain any future problem that may come. Businesses have been told this for the last 400+ years. “This time it is different.” This is far from the truth. All you have to do is look at the last U.S. recession to see the fallacy in this statement and the fact is that you are being misled by governments and media reporting. I am reminded of Earnest Hemingway’s question – “How does one go bankrupt?” “The answer is slowly, then all at once.” The method we use at Data Stats indicates “slow” is the description of what is happening now in the U.S. The United States Bureau of Economic Analysis (BEA) begs to differ. They say the economy is going at high speed. The economy is growing, but not as fast as the “powers” say. Some, and me included, feel that the U.S. economy is just above stall speed.
Seasonally Adjusted GDP Data
For the last twelve years we have been using the Year to Quarter Gross Domestic Product (GDP) growth data to measure the state of the United States and Canadian economies. Both governments advertise a seasonally adjusted quarter to quarter growth rate. They do publish the raw data but they bury it in their reports. You have to look for it. The problem with seasonally adjusted GDP estimates is that they assume businesses will make the kind of investments necessary to continue the growth. A 4.2% increase in quarterly seasonally adjusted GDP, as was the case for the U.S. second quarter of 2018, could very well be the peak and everything else goes downhill.
Year To Quarter GDP Data
Both sets of data we have been discussing are estimates. Neither is perfect. However, we believe Year to Quarter data is much more accurate in predicting the movement and the end of the year result. If you are looking six to twelve months ahead, the end of year result is what you need.
The table below compares actual final United States’ GDP to what is predicted by averaging the four quarters of seasonally adjusted data as put out by the U.S. Bureau of Economic Analysis and the unadjusted Year to Quarter data also put out by the BEA:
Comparison of Seasonally Adjusted and Unadjusted U.S. GDP Growth Rates
| Year |
GDP for the
Year (%) |
Ave. of Four Quarters
Seasonally Adj. GDP (%) |
Ave. of Four Quarters
Unadjusted GDP (%) |
| 2009 | -3.5 | -0.5
| -2.4 |
| 2010 | 3.0 | 3.2
| 2.9 |
| 2011 | 1.7 | 1.6 | 1.8 |
| 2012 | 2.2 | 1.9 | 2.1 |
| 2013 | 1.7 | 1.6 | 1.9 |
| 2014 | 2.6 | 1.3 | 2.3 |
| 2015 | 2.4 | 1.9 | 2.4 |
| 2016 | 1.3 | -0.4 | 1.6 |
| 2017 | 2.3 | 2.4 | 2.3 |
Average Deviation between Actual and Seasonably Adjusted GDP=0.65%
Average Deviation between Actual and Non-adjusted GDP=0.22%
Seasonably Adjusted Error is 2.9 Times that of Non-adjusted Data Error.
As you could see the average for the seasonally adjusted data is off by 0.65%. For the unadjusted data it is 0.22%. The average seasonally adjusted data error is 2.9 times greater than the unadjusted data.
Small Errors Have Consequences
In 2017, the United States economy was US$19.39 trillion. A 1% error equals US$193.9 billion. A 0.65% error equals US$125.45 billion. If you are a small to medium size business this is a huge amount. It is certainly a significant value. Over the past nine years, the error (difference) rates for seasonally adjusted quarterly rates range from as high as 3% to a low of 0.1%. The Year to Quarter error ranged from a high of 1.1% to a low of 0.1% and predicted the exact yearly rate twice, once in 2015 and again in 2017. I believe that this supports our feeling that the raw data is the preferable measure.
GDP is a snapshot in time. It does not tell us where the economy is going. You need to use the numbers from several quarters to develop a trend. Three numbers are the minimum required. The more readings utilized, the more confidence you have in the trend. The confidence in the trend is also a function of the accuracy and the variation in each of the data points. For example, if a data point contains a 0.65% error, it may give you a false or inaccurate trend. You do not want to base your decisions on this type of information. I believe that this is another weakness of seasonally adjusted data.
What Is The Economy Doing Now?
The raw data tells you what the economy has done and what it is doing now. Seasonally adjusted data tells you what the economy may do. It may or may not predict the future accurately. I believe the unadjusted data will more correctly predict the future.
If you have any further suggestions, do not keep it to yourself. Help your fellow readers!
If you have any questions, comments or suggestions drop me a line at rdm@datastats.com.
|
