Well, one thing I can say is that is not pretty, but i am giving it to you anyway.  I tried to get you a graph so you could visualize the data, but i only get into Powerpoint charting periodically and it seems like every time i do, i have to learn everything back over again.  I did get a visual myself from my Excel file, but then it would not copy with the data--so i don't know.  

It really does not matter.  The data speaks for itself.

Now here is what i did.  I got my Medicare costs numbers first.  And i checked them out from a couple of different sources.  I learned several different things about the Medicare program in the process--like its money sources, like the percent between medicare and medicare advantage, percent that goes to hospital care versus physician clinics versus retail prescriptions (telling me that those numbers could be looked at in the same way to see if you can recognize anything there, too.

Then i got my CPI numbers and 65+ population numbers for the same period (2000-2021)

Now you have to understand that i took a few "shortcuts" with the data.  What i mean here is that sometimes i simply found a couple of solid spots within the data, then prorated the growth between those two periods.  (If you have any questions about that just ask, but i think it is totally justifiable based upon the fact it would not incease the say "error range" of my analysis by that much.

For example, I had a good number for 2000 and 2005 for medicare, but not the inbetween numbers, so i prorated the growth linearly between those two periods.  (Yes, i know i could probably find the real numbers, but for what i was trying to do, i didn't think it would matter.  Just the same, i could have done more digging for actuals, i just didn't want to spend the time.  I did not think it was necessary.  All the medicare numbers after 2011 are actuals, however, and not prorated.

So the following data shows is this.  If you took the costs as they were in the year 2000, then add the effects of cpi and the increasing population of those 65+ (btw, the baby boomer population began hitting medicare in 2011 and is reflected in the numbers) and simply based upon that you would expect medicare costs in 2021 to be $564.7 Billion, while the actual medicare costs in 2021 were $839.3 Billion.  Again based on 2000 numbers.  You could always choose another starting point, or do a whole bunch of different starting points.  The "actual cost growth rate from 2000 to 2021" was 278% whereas the "expected cost growth rate from 2000 to 2021" was 155%

Now it is that difference that needs to be explained.  We have looked at both cpi and increasing aged population growth--what else made these numbers so much higher than otherwise expected?  There may be some legitimate reason for some of it.  (For example, you can see quite a bit of the effect that covid seems to have had on at least one of the numbers.  But that's an outlier, and really shouldn't be considered a valid point in the overall aspect of the analysis itself.

One thing that we would probably want to look at is whether health care benefits changed or look to see if we can actually see any signs of unexpected "health care improvements" that would also justify increases, but that would just be a part of the next level of analysis.  I doubt if anything like that, however, will explain the entire difference.

One thing i found really quite funny.  I noticed someone had done analysis that showed that Medicare cost increases tracks right along with private health care increases--somewhat implying that medicare is run as well as the private programs.  The funny thing is this. "Both" medicare and the private health care costs have been going up faster than you would otherwise expect.  Your comparing apples to apples--and everyone is getting fucked (just my opinion)

Anyway, take a look.  It probably took me about three hours in total--fiddling around getting the data, doing some prorating, trying to graph it in powerpoint, etc.  Would i stand by it 100% right now at this moment?  No, but i am pretty confident that my calculations and data are correct, but typically, i would sit on this for a few hours or a day, trying to think what i could have done wrong.  As it is, the results came out somewhat how i would have expected them to based upon what i know about health care costs.  Actually, when i first ran my analysis i had medicare expected and actuals tracking right along with each other--all the way through.  I was surprised, but i graphed it and looked at it that was my analysis was showing--until i discovered that one of my Excel formulas was all fucked up.

I do not know about you, but i have not seen anywhere in my limited research where medicare costs have been looked at in this manner above.

ps.  I am planning to go "Dark", starting Tuesday.  And this go around i think i can actually do it.

ps2.  I will show this around to a few of my people and see what they think.  In the meantime i might do a little more digging to try to validate my numbers, but unless i find something, right now i am standing by this.

ps3. Like i said, it's not too difficult to do--maybe even the Government could "proof" my numbers and analysis.  They could put out an rfp (that would probably only take 3 months), then they could solicit bids (another 3 months), the new contractor then can spend a year trying to see whether my analysis is correct or not.  Which by the way it isn't, because i didn't bother getting "every exact number" that i could have or should have gotten--and that would be in the new contractor's report, too--showing you how much better they are than i am.  Go figure.