Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

On the graph they shared earlier, 2020 is 44.82%, exactly on the 1960 - 1992 trendline. The 2022 and 2009 dropoffs coincide with major disruptions (don't know about the late-90s one), which almost seems to hint that's the value we should be looking at: +10.57% in 20 years.


2019 was 35.97, and 2022 was 36.26, but we shouldn't use those, we should use the 2020 number? Come on, the 2020 and 2021 numbers are not representative of the overall trend.


Sure but the 1990s, after Clinton's federal worker bloodbath and other expenditure cuts, is also an outlier and by the same argument you could say it is not representative of the overall trend. You can't just pick and choose your preferred outliers and call the inconvenient ones not representative.

Covid spending was an "outlier" but it is not an uncommon kind of outlier when there is some market crash or trillion dollar wars etc -- there was almost as large a spike, and a broader peak meaning more overall expenditure as a proportion of GDP only 10 years before Covid.


> Sure but the 1990s, after Clinton's federal worker bloodbath and other expenditure cuts, is also an outlier and by the same argument you could say it is not representative of the overall trend. You can't just pick and choose your preferred outliers and call the inconvenient ones not representative.

I think it is perfectly fine to say that the 90s are not representative of the overall trend.

I'm not cherrypicking. I'm not saying it's actually going down. My argument is very simple, that "consistently up" is not a correct description of the situation.

If you look at it a couple years at a time, it's swingy and there's no real pattern at this level. If you look a decade at a time, it went up quickly for quite a while, it went down more slowly for a while, and now it's going back up even more slowly and it's still below the peak. If you zoom out really far, then things are close to flat over the last 40 years.


I didn't see using the word consistently, so it seems like you're attempting an escape hatch with this strawman. I called it steadily upwards over the past 100 years until recently when it's becoming more erratic but still looks to be increasing. Which is the simple truth. I can't understand how so many people in this thread are struggling so badly with that.

And the trend is upwards, in the past 40 years. Run an analysis on it and it's trending upwards. It's not close to flat, it's increasing by a couple of %, i..e., the spending ratio is increasing around 5% in that time. Which is enormous. You can cry foul that's due to covid and the housing bubble spending, but that's reality, that's the money that is getting spent and added to the debt and is what people are concerned about. Even if there was no Clinton cuts outlier going the other way, you still couldn't discount those spikes as outliers and therefore not representative. Because they exactly represent the reality of the situation. This is not a concern about a measurement error or sample that's not represenative of a population, it is the actual money that is being spent. If there weren't these big outliers for wars and stimulus and everything else then sure, things would probably be better. But there are.


> I didn't see using the word consistently, so it seems like you're attempting an escape hatch with this strawman.

Are consistent and steady not synonyms? You seem to have been arguing consistency, I'm not trying to make a strawman here.

Are you saying it's steady but inconsistent?

> Run an analysis on it and it's trending upwards.

Pretending everything can be coerced into a straight line is statistics abuse 101.

It also forces any data set ever to be "steady".

2022 is lower than 1982. That's also overly simplified, but it helps show that trying to cram things into a single number is not correct here.

> you still couldn't discount those spikes as outliers and therefore not representative

I wasn't discounting any spikes other than covid. The 2008 spike spent most of its time below the early 90s.


> Are consistent and steady not synonyms? You seem to have been arguing consistency, I'm not trying to make a strawman here.

> Are you saying it's steady but inconsistent?

You don't have to ask all these obtuse rhetorical questions, you can just read exactly what I'm "saying".

> Government expenditure as a proportion of GDP has been rising steadily for a century and is now approaching 40%. From IMF: https://www.imf.org/external/datamapper/exp@FPP/USA. The recent trend is shaky due to Covid, housing crash, maybe dot com crash, etc., but it looks like it's probably still trending upward.

Point out what is so wrong about that, and I will point out an insufferable nitpicker who is incapable of having a substantial discussion but they're upset about something so they feel they simply must score internet points with idiotic arguments about semantics (which they are wrong about anyway).




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: