Posted by
Patrick Bohan on Tuesday, November 10, 2009 9:30:03 AM
There is a fundamental flaw with how the Congressional Budget Office (CBO) projects national and legislative budgets. Most people think that accuracy and precision mean the same thing. They do not! For instance, the CBO projections for health care costs are precise, but they do not take into account accuracy. The CBO only shows the accuracy, or error, of their budget projections once the legislation matures over time. By this time it is too late correct legislation and policies that have gone over budget.
Here is an analogy from my engineering background to distinguish the difference between precision and accuracy. I spent a lot of time working on analog to digital converters (ADC). The precision of an ADC can be measured by the number of bits it has. For example, a 10 bit converter is more precise than an 8 bit converter. In fact, a 10 bit converter is four times more precise than an 8 bit converter. On the other hand, accuracy measures the error of the converter, and most ADC’s have an accuracy of +/- ½ bit. Thus, the 10 bit converter and the 8 bit converter have the same level of accuracy, or error. Another way to look at accuracy is how repeatable a result is. If an ADC yields the same exact result to a specific input when temperature, humidity, test fixtures, and other variables are changed, the ADC is indeed accurate.
The same analogy applies to the federal government projecting budgets. For instance, a few months back the Obama administration announced the federal deficit will increase from 7 trillion to 9 trillion over the next 10 years. The increase in 2 trillion did not happen because our government just spent 2 trillion more dollars. It happened because the variables they used to calculate the 7 trillion dollar deficit have changed and the budget is now 2 trillion dollars higher. More recently, the CBO projected what the health care proposals from congress would precisely cost the American public. For example, the bill by Max Baucus came in at precisely 829 billion dollars. To come up this calculation, the CBO had to make some assumptions as to what values to use for all variables in the equation. For example, the cost of health care reform would vary depending on the rate of inflation, the value of the dollar, the unemployment rate, the obesity rate, the poverty rate, population growth, and a plethora of other variables. If the CBO obtained a precise budget of 829 billion using a rate of inflation of 3.5%, the accuracy of this result will vary if the inflation rate is higher or lower. Considering the government has recently flooded the economy with nearly a trillion dollars, one could expect the rate of inflation to be 5 or 10% in the coming years. If this did occur, the 829 billion dollar projection over 10 years would actually be 1 trillion or more dollars. Hence, the CBO projection would be off by over 20%, meaning the accuracy of the 829 billion dollar projection would have an accuracy of +20%.
Another important thing to note, very rarely do government budgets go under their cost projections, they are generally much higher. Thus, the CBO usually makes it projections using best case numbers. This is why it is imperative for the CBO to also put together a worse case scenario for their budget projections. For instance, the CBO numbers for the Baucus health care bill may cost anywhere from 829 billion to 1.2 trillion using best case and worst case numbers. Using these number extremes we could therefore conclude the cost of the bill would be around 1 trillion dollars with an accuracy of +/- 20%. This would be a much better way to project budgets instead of trying to cover them up with rosy and unrealistic targets. Until the CBO, and the government, implement accuracy into their projections for budgets, they cannot be trusted. And we will therefore continue to run up our national debt.
My Book: Is America Dying? (Amazon.com, Barnes and Noble)