So, this is really random, but I was making my reply to Hybred and I was looking in spreadsheet a noticed something that seems weird:
https://www.cbo.gov/about/products/budget-economic-data
Download the spreadsheet for the long-term projections under Jan 2017, and go to the tab titled supp table 1, and go to the columns for revenue, outlays, and GDP. I was curious what the long term growth rates they used for these were so I did a formula to see what the compounded annual growth rate was comparing the 2047 numbers to the 2017 numbers. The formula for that is, =rate(30,0, negative amount for 2017, positive amount for 2047). I got 4.41% for revenues, 5.16% for outlays, and 4.04% for gdp. That tells you, if you were to have that growth rate every year, you would get the ending value by 2047.
Now, if you look at the growth rate in each individual year, it is different. IE, for revenues in the year 2018 the growth rate is 5.88%, for 2019 it's 2.78%, and so on (comparing the growth from 1 year to the next). For each year, it's different.
Here's the really weird part. If I take all the simple average growth rates for each of the years, they equal exactly the same as the compounded growth rate. 4.41% for revenues, 5.16% for outlays, and 4.04% for gdp. If you know about the difference between simple and geometric averages, that is fucking weird. These numbers usually aren't the same. For example, took stock market returns over a period of 30 years, the compounded rate of return is not going to equal the simple average of the annual returns for each year. This is because the percent growth of any given year is calculated off of a different base amount than the growth rate of the other years.
I thought this was interesting and it could mean a few things. One, it could be a complete fluke. Two, it could have to do with the way the model was done. Maybe the model uses the assumption that the simple average of the annual growth rates should equal the compounded average of the growth rate for the entire period because some evidence shows this yields a better overall prediction. Three, this cut have been a big shortcut in the model and possible evidence of building the model to show a certain end result. IE, if you want to have it show that the deficit will be this big by the end of a period, then this is what you have to do to get there.
The thing that really makes me stop and think here is that the most significant parts of the budget, individual tax revenues, social security and disability, and medicare should all be based on actuarial data about life spans, when people stop working, when they get disabled, when they start taking social security, and when they die. This calls into question whether shortcuts were taken on any of that which may compromise the underlying assumptions of the model. Of course, I don't really see evidence for that, except for the fact that it seems weird that the model would be based on all of those actuarial assumptions and then conform to a very arbitrary rule about growth.
Or yeah, maybe it doesn't mean shit and it's only interesting to me because I'm a fucking wierdo. But I thought I'd share anyways.
https://www.cbo.gov/about/products/budget-economic-data
Download the spreadsheet for the long-term projections under Jan 2017, and go to the tab titled supp table 1, and go to the columns for revenue, outlays, and GDP. I was curious what the long term growth rates they used for these were so I did a formula to see what the compounded annual growth rate was comparing the 2047 numbers to the 2017 numbers. The formula for that is, =rate(30,0, negative amount for 2017, positive amount for 2047). I got 4.41% for revenues, 5.16% for outlays, and 4.04% for gdp. That tells you, if you were to have that growth rate every year, you would get the ending value by 2047.
Now, if you look at the growth rate in each individual year, it is different. IE, for revenues in the year 2018 the growth rate is 5.88%, for 2019 it's 2.78%, and so on (comparing the growth from 1 year to the next). For each year, it's different.
Here's the really weird part. If I take all the simple average growth rates for each of the years, they equal exactly the same as the compounded growth rate. 4.41% for revenues, 5.16% for outlays, and 4.04% for gdp. If you know about the difference between simple and geometric averages, that is fucking weird. These numbers usually aren't the same. For example, took stock market returns over a period of 30 years, the compounded rate of return is not going to equal the simple average of the annual returns for each year. This is because the percent growth of any given year is calculated off of a different base amount than the growth rate of the other years.
I thought this was interesting and it could mean a few things. One, it could be a complete fluke. Two, it could have to do with the way the model was done. Maybe the model uses the assumption that the simple average of the annual growth rates should equal the compounded average of the growth rate for the entire period because some evidence shows this yields a better overall prediction. Three, this cut have been a big shortcut in the model and possible evidence of building the model to show a certain end result. IE, if you want to have it show that the deficit will be this big by the end of a period, then this is what you have to do to get there.
The thing that really makes me stop and think here is that the most significant parts of the budget, individual tax revenues, social security and disability, and medicare should all be based on actuarial data about life spans, when people stop working, when they get disabled, when they start taking social security, and when they die. This calls into question whether shortcuts were taken on any of that which may compromise the underlying assumptions of the model. Of course, I don't really see evidence for that, except for the fact that it seems weird that the model would be based on all of those actuarial assumptions and then conform to a very arbitrary rule about growth.
Or yeah, maybe it doesn't mean shit and it's only interesting to me because I'm a fucking wierdo. But I thought I'd share anyways.