An FI Difficulty Index

[black_son_of_gray]

So I've been thinking about how @Jacob sometimes refers to high and low incomes as being 'easy-mode' and 'hard-mode', respectively. I figure it would be informative to quantify this because a lot of the people on here in 'hard-mode' might be discouraged with their relatively lower savings rates, but really - they are quite impressive! (Or said another way, some of the high income, high % savers need a dose of reality )

So I came up with a -100 to 100 scale… the more negative, the 'harder' saving is; the more positive, the 'easier'.

The FI difficulty index:
(% rank of personal income) - (saving rate %)

What immediately jumps out is that if both numbers are the same, then the index is 0, so a 50%ile income person saving 50% is 0, etc.

The calculation also gives rise to equivalencies across incomes (that's the whole point, actually). For example, a 50%ile income person saving 75% is about as 'difficult' as a 25%ile income person saving 50% (both -25, or what I will arbitrarily call 'medium hard')

Quite simple, really. Maybe too simple, I was going for a BMI-like calculation that is super easy but relatively accurate - If anyone can think up some other or better way to quantify this aspect of FI, please share!
(US income numbers if you're curious, look at Percent Below)
For the record, I am on 'medium easy' mode, at ~ +25.
It would be great to see who the hard mode folks are, so we can celebrate them for the wonders they are!

[theanimal]

This is cool. Somehow factoring COL would be neat too. Someone who saves 80-90% living in Manhattan could possibly be more skilled and have a more difficult time than someone saving the same amount on the same income living in Topeka, Kansas.

I'm -56.

[henrik]

-19 (US numbers; less the fact that the Wikipedia article probably shows gross income)
+43 (global)

[jacob]

@OP - Great idea. I'm quite impressed for developing a parsimonious measure. I wish I had thought of this!

This is a very good way to measure one's ability to stretch beyond one's peer group.

Here's a quite reference: http://politicalcalculations.blogspot.c ... ntile.html

I was -30 during grad school and -10 during postdoc (academic bench work).

Since I was FI shorty after that, my savings rate was beyond 100% of "earned" (W-2) income, so this metric gets kinda iffy, because technically my savings rate relative to income has been beyond 100% since that.

[JL13]

Am I understanding this correctly? So if 77% are below you (you're closer to the top than the bottom) and your savings rate is 50%, then your ERE score is 27?

Is this posttax I imagine?

So it's like golf then, a lower number is better?

Should we create a Wheaton scale then?

[black_son_of_gray]

@ J_L13 ... Yes, in your example, the score would be +27.

As for posttax or not, I was thinking initially "Well, the income distribution is heavily skewed, so maybe this isn't accurate across the whole range" - which is probably true. But if we consider that higher income people, who are in that long tail, are also taxed more heavily (maybe not depending on source), then maybe that serves to balance things out. Honestly, I don't know.

As for 'better' - in your example, +27 would just indicate relatively easier conditions for achieving FI if one was dedicated to doing so. Whether the score is positive or negative doesn't mean better or worse, per se, just degree of difficulty at that time. I get the sense that people may start their journey all over the scale at different difficulties, but as they build their assets/resources, they all end up with an escape velocity that continually shifts them towards easy. (Or becomes meaningless, as Jacob mentions. I think this is very much an accumulation-phase metric).

That's an interesting idea about Wheaton scales - I wonder how big of a spread would lead to a barrier in understanding/communication, and how often that manifests in the forum responses. Can a +50 and a -50 really relate?

[henrik]

Can a +50 and a -50 really relate?

Note my two very different numbers in the post above depending on background data. I relate to myself quite well:)
I think the main problem with interpreting results from this is that the income distribution dataset you choose determines much more than your savings rate. Maybe instead of national numbers it would make more sense to go by city or neighbourhood? Or education or profession? Maybe the global distribution is the only correct one, because after all, you could move and thus make it easier or harder for yourself, and choosing not to move is part of how easy or hard FI is? I don't know...

[rube]

And should we use a correction factor for kids and/or a SO?

[JasonR]

o

[black_son_of_gray]

@henrik: fair point about relating to oneself, but I think there is some value in calculating different numbers as long as the context of the dataset is considered. If two people were comparing their numbers, it would make sense for them to be in the same reference frame, and therefore use the same dataset that is most relevant to them. To return to the example, on a global scale, can a +50 and a -50 really relate? Many on this forum would be close to the +50 mark. What would a -50 be? A child laborer in Bangladesh? I don't know, I'm making that up - but I know I can't comprehend their mindset or understand their problems very accurately from my position. But you are correct about dataset choice, and that could be a problem with this international crowd.

@rube: Not sure - how much savings % does a kid or SO eat up? I can definitely understand that dependents make FI more difficult, but that seems like a rabbit hole of questions that center around what is 'necessary' (e.g. paying for their college could be a huge expense making FI more difficult, but is it absolutely necessary as a parent? Not every country/state has crazy college expenses, etc). For some people, children are a very small additional expense, for others they are a huge expense - it's not easy to parse out what is skill and what is less controllable factors. Maybe the best thing to do is remove child-care costs from the income side of things instead, if that can be done cleanly and accurately.

[jacob]

Instead of savings rate, use (income percentile / spending percentile). This will solve many of the problems above.

Assuming that it's normal to spend 100% of income, the same table can be used. E.g. if you make 40k and spend 10k, the number would be 65%/18%=3.6

Higher numbers are more impressive.

[JL13]

I just put it all in a spreadsheet for easy reference....but the chart is so huge you can't even read it.

http://i66.tinypic.com/339pw7d.jpg

[theanimal]

Doesn't that just place a higher emphasis on income? And favor those that earn more?

[jacob]

It favours those who earn more in the linear sense (x*) while it favours those who spend less in the harmonic (meaning 1/x* or leverage). The highest score will be high income and low spending, so my measure gauges deviation from peer group --- the more you make, the more potential you have to deviate. Low scores will be for people who earn and spend around the median of everybody.

[JamesR]

You could factor in COL differences between regions by focusing on "local income percentile" perhaps. So a salary of $40K should have a different income percentile depending on if you're in Manhattan or Topeka.


A correction factor for kids and/or a SO is perhaps impossible to do. What about dual incomes and no kids? HUGE advantage over the solo income individual. Could possibly come up with some kind of formula that's based on economies of scale, but everyone would disagree on what the marginal cost of dependents should be.

[henrik]

Correction factors based on life choices are tricky here. Should we also correct for having an expensive car or being "bad at math"? Seems like it would defeat the purpose at some point.

[BlueNote]

This is a good idea. I sometimes like to put myself in a 'league' and compare myself to others in the same league (30 something Toronto DINK operating around the middle of the local gross income average). Ultimately though there is something to be said for being totally mathematically honest. I could potentially negotiate a work form home arrangement and move to a cheap Canadian town and save even more (cut the rent in half). Heck I could probably arrange to work from Chiang Mai or Agentina most of the year if I wanted but I don't think my wife could. There's also something to be said for being in a relationship with someone who isn't ready/willing to fully adopt FIRE at the level you are. I am having to sell it in small bits due to the contrast of moving from one 'wheaton' level to the next. Essentially I am slowly boiling a frog here

[jacob]

Correction factors based on life choices are tricky here.

The most complicated accounting for life choices is also known as the .... tax code. It can be done but...