Author Topic: Towards a Unifying Theory of Math is Math and Behavioral Economics  (Read 16301 times)

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
This post is a tip of the hat to @boarder42 and some of the issues he's raised.  It also raises what I believe are the fundamental questions for this board: when do we trust math, and when do we concede to behavioral economics?

To some extent, Mustachianism is about overcoming behavioral irrationalities and applying correct rational analysis in the case of marketing, advertising, etc.  Do I really need that car?  Will the fancy vacation bring me happiness?  We are in the minority because it takes concerted effort to overcome the behavioral tendencies that make us want to do what everyone else is doing, buying, etc.

Boarder42 applies that analysis dogmatically with respect to pre-paying a mortgage (at least at locked-in low interest rates for an extended term).  Over a 30-year period, the market has always had greater returns than the 3% or 4% that current mortgages are at, so from boarder's perspective, it's just a math problem.  The math says invest in the market over pre-paying, so that's what you should do, no exceptions.  If you want to pre-pay your mortgage to feel more secure, etc., that's fine, but understand that you're making a bad math decision and at least know the true cost of your irrational desire for security.

Although the tone is sometimes a little strong, his math analysis is spot on, and it's consistent with the analysis MMM and the forum often dictates when someone wants to buy a new fancy thing.  How much will that cost you compared to investing, and is it really worth it to have fancy car, boat, etc. for that cost?

So far, so good.  But I have been more agnostic about pre-paying the mortgage, because I think we often have trouble with math is math, and the question is not whether we can see math is math, but are we really going to act consistently with that over the 30-year period.  To the extent that people are risk averse (most are, per behavioral economics), what if they end up acting consistently with that risk aversion and try to market time or whatever else that is not the rational response, thus making them unable to successfully live up to the math is math principle. 

I think you see that when you consider whether people will take a loan out on a paid-off house to invest in the market.  Even though it's the same principle as not pre-paying the mortgage, I think most people would view it as very risky.   We'd need data for that of course, but if so, it highlights the impact of behavioral economics.  If the framing of the question causes people to give a different answer, then we have to be skeptical whether people are going to act rationally at all times.

That takes me to asset allocation questions.  Generally, the responses I've seen on this forum when people say that they're nervous about the market are (1) trust the market, it goes up over the long term despite bumps, and (2) if you're nervous, that means your asset allocation is off and you might need to dial back your stock investments to match your risk aversion.

The first is the rational response.  The second is not.  The answer to the second under the math is math principle is that you need to get comfortable with the math and internalize it, not act irrationally by dialing back your risk exposure because you're irrationally uncomfortable with risk.  Yet I don't see the same level of criticism of the response that people should change their asset allocation to match their risk appetite.  Why?  Assuming a person with a 30-year or more time horizon for money (not someone close to retirement who might need a higher bond allocation, for example), the answer under the math is math principle is that the person should do work to become more rational, not change their investments to match their irrationality.

Which takes us to market timing, market efficiency, and the 4% rule.  The overwhelming advice on this forum is to trust the market because we're not smarter than the market.  Buy index funds and don't market time.  That's rational and correct based on the data.  Yet up comes the bitcoin threads, and suddenly the vast majority of posters ridicule the bitcoin price for whatever reason is obvious to them that bitcoin is just the tulip market.  But what about market efficiency?  What is the informational advantage that these posters suddenly have over the bitcoin market that they don't have over the stock market?  If the bitcoin price is "obviously" wrong to a random MMM poster, why has the market not priced that in?  If it's equally "obvious" to another random MMM poster that the stock market is overvalued, why is that person wrong and doesn't know more than the market?

The bitcoin problem highlights the principle that we start making errors in rational analysis as soon as we get outside our comfort zone.  There might be a 1% chance that bitcoin is going to be worth a gazillion dollars, and a 99% chance it's worthless, and the market has potentially priced that correctly.  But because we don't "understand" that math, we dismiss it, and therefore give an irrational assessment of what could be a perfectly efficient market and price.

All of which raises the question to me of how far we take the "math is math" principle, and when do we concede that people are likely to revert to their natural behavioral tendencies.  With respect to risk aversion, everyone talks a good game about never selling stock, and it's been easy to hold that view since 2011 or so.  But if we're being honest, how many of us had nearly as much money at risk then as we do now?  People statistically buy high and sell low because that's behavioral economics (i.e., human nature), but we're all confident we're the exceptions because we understand rationally what to do.  Maybe, and it's worth aspiring to.  But I've just noticed in these threads the many different responses to analytically similar problems based on how the question is worded, and I wonder just how much we can count on overcoming our behavioral tendencies and acting correctly and perfectly rational based on math is math principles.

Again, props to boarder42 for triggering some of these questions.  I'm intrigued by how we reconcile the responses with respect to these issues.  I'm also going to mention @maizeman because I find his responses to be thoughtful and well reasoned and have incorporated some of his terminology into the framing of these issues.  If I did so incorrectly, that's on me!


TexasRunner

  • Pencil Stache
  • ****
  • Posts: 926
  • Age: 32
  • Location: Somewhere in Tejas
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #1 on: January 11, 2018, 07:37:47 AM »
I think one of the biggest things is the Investment Policy Statement, how accurately its worded and how systematically it is applied...

I would also bet that most on MMM do not have one.

I would also bet that most who do have one do not mention spending, leveraging or mortgaging specifically in the IPS, which is also a big deal...

Very good post and questions raised, I look forward to following this conversation.

terran

  • Magnum Stache
  • ******
  • Posts: 3796
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #2 on: January 11, 2018, 08:16:23 AM »
First of all, great post! Super interesting.

Which takes us to market timing, market efficiency, and the 4% rule.  The overwhelming advice on this forum is to trust the market because we're not smarter than the market.  Buy index funds and don't market time.  That's rational and correct based on the data.  Yet up comes the bitcoin threads, and suddenly the vast majority of posters ridicule the bitcoin price for whatever reason is obvious to them that bitcoin is just the tulip market.  But what about market efficiency?  What is the informational advantage that these posters suddenly have over the bitcoin market that they don't have over the stock market?  If the bitcoin price is "obviously" wrong to a random MMM poster, why has the market not priced that in?  If it's equally "obvious" to another random MMM poster that the stock market is overvalued, why is that person wrong and doesn't know more than the market?

The bitcoin problem highlights the principle that we start making errors in rational analysis as soon as we get outside our comfort zone.  There might be a 1% chance that bitcoin is going to be worth a gazillion dollars, and a 99% chance it's worthless, and the market has potentially priced that correctly.  But because we don't "understand" that math, we dismiss it, and therefore give an irrational assessment of what could be a perfectly efficient market and price.

I do have one quibble with this part. I don't see anyone (at least no bogleheadish indexer) suggesting the market is efficient so that means it would be totally rational to invest in a single stock. The bitcoin market is simply multiple places to buy and sell a single investment, while "the market" is multiple places in which all publicly traded investments are available. You most efficiently invest in (an approximation of) "the market" by buying a few broad index funds. While bitcoin may be priced rationally in the sense that a 1 in 10 chance of getting $10 is worth anything less than $1, you still have a 90% chance of getting no dollars.

I also think that bubbles (whether or not bitcoin actually is one) actually prove your point that we are not purely rational math-is-math beings. Bubbles are the natural result of people seeing success and notoriety and wanting to get in on that regardless of whether the underlying investment is sound. A bubble is the breakdown in the efficient market, so if we accept that we can't personally identify bubbles, we should invest in the whole market so at least the bubbles we don't identify don't wipe us out when they return to efficiency (pop).

I would also suggest that in a pure math-is-math sense we can actually identify bubbles if we look at the underlying assets we buy with an investment. When an underlying asset will provide a future income stream that is worthwhile based on the initial investment then it's not a bubble, when it doesn't then there is at least some bubble going on in the form of speculation. That speculation might pay off or it might not, and the bubble might pop or it might slowly deflate, which is the difference between losing "everything" and slightly underperforming. We're all speculating (even government bonds have some small chance of failing to deliver their expected returns if the government becomes insolvent), so the best way to help avoid losing everything in these speculations is to diversify.

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #3 on: January 11, 2018, 08:45:13 AM »
Thanks for the kind words Cycling Stash!

I do have one quibble with this part. I don't see anyone (at least no bogleheadish indexer) suggesting the market is efficient so that means it would be totally rational to invest in a single stock. The bitcoin market is simply multiple places to buy and sell a single investment, while "the market" is multiple places in which all publicly traded investments are available. You most efficiently invest in (an approximation of) "the market" by buying a few broad index funds.

I think you're using the word "efficient" to mean something different than the way the word is used in the efficient market hypothesis. I think any bogleheadish indexer would agree that the expected value of identifying "underpriced" stocks and buying then, and identifying "overpriced" stocks and shortselling them doesn't exceed the expected value of just buying a broad index. A simpler way of saying the above is that the pricing of stocks is efficient.

Not to single you out, I think the problem a lot of people run into is that thinking an argument over whether or not markets for $X are efficient is the same as the argument for whether or not it makes sense to buy $X as an investment. However, the reality is that the pricing of $X is efficient doesn't necessarily imply that $X is a good investment. It just means there are no apparent opportunities to make free money through arbitrage or other trading strategies.

A good example of this is commodities. The price of most widely traded commodities are generally efficient although just like stocks that doesn't prevent people from speculating and trying to outguess the market. That said, I would highly recommend you NOT invest a significant chunk of your stash in porkbellies, because the expected return on porkbelly futures is zero (sometimes you'll make money, sometimes you'll lose money and over the long term the two will balance out), and the expected return on delivered porkbellies is negative (you'd have to pay a big electric bill to turn your storage space into a freezer, pay rent on that storage space, and people are going to pay a lot less for a five year old porkbelly than a fresh one, even if it's been kept properly frozen).

That takes me to asset allocation questions.  Generally, the responses I've seen on this forum when people say that they're nervous about the market are (1) trust the market, it goes up over the long term despite bumps, and (2) if you're nervous, that means your asset allocation is off and you might need to dial back your stock investments to match your risk aversion.

The first is the rational response.  The second is not.  The answer to the second under the math is math principle is that you need to get comfortable with the math and internalize it, not act irrationally by dialing back your risk exposure because you're irrationally uncomfortable with risk.  Yet I don't see the same level of criticism of the response that people should change their asset allocation to match their risk appetite.  Why?  Assuming a person with a 30-year or more time horizon for money (not someone close to retirement who might need a higher bond allocation, for example), the answer under the math is math principle is that the person should do work to become more rational, not change their investments to match their irrationality.

When I've seen that advice I've always read it as "if you're already nervous now, there is a significant risk you're not going to be able to stick to your asset allocation in a crash, and the losses from buying stocks high and selling them low are greater than the losses from a constant suboptimally high allocation to bonds." Essentially a rational response to a particular person's recognition that they won't always be able to force themselves to make rational decisions in some future scenarios.

If people can figure out a way to change their mindset to the point that it changes their future potential behavior, that's a financially preferable option to changing their asset allocation. (Changing mindsets on that fundamental a level is hard, but at least a fair number of people can manage it.)

If people think it's a good idea to shift into higher bond portfolios just to avoid FEELING nervous I agree that's incredibly bad advice.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #4 on: January 11, 2018, 08:52:40 AM »
"Behavioral Economics" is just code for "bad math".

The beauty of math is that it is correct.  There is no arguing with it.  Your feelings don't matter.  If there is a mathematically correct way to do something, and you choose ANY other way for ANY other reason, then you are wrong. 

Which is fine.  Humans make deliberately wrong decisions all the time.  But we don't need to pretend that our wrong decisions are actually right decisions just to make ourselves feel better.  We don't need to resort euphemisms like behavioral economics.  Just admit you've deliberately made suboptimal choices because you're bad at math, and we can all move on.

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #5 on: January 11, 2018, 09:23:04 AM »
"Behavioral Economics" is just code for "bad math".

No, behavioral economics is the study of why human beings made decisions, particularly when those decisions aren't the ones rational optimization of utility ("the math") would suggest.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #6 on: January 11, 2018, 09:37:56 AM »
To me if your behavior is irrational with regards to this and you cannot make sound decisions based on the math and follow your ips then you likely would benefit from an advisor who can help you. To compound irrational investing by intentionally skewing an asset allocation or paying down a mortgage to feel better would be a giant red flag that you likely need an advisor and would benefit from their ability to help you ride out down markets without selling. Bc if a mortgage payment pre fire makes you nervous how will you handle drawing down your stache in fire in a down market.  And increasing bonds past 20% is a poor idea based on historical math in most cases outside of doing a reverse equity glide path. But if a person plans to do that they've evaluated alot of math around fire pretty deeply.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #7 on: January 11, 2018, 09:48:45 AM »
The beauty of math is that it is correct.  There is no arguing with it. 

Math itself is always correct but garbage in still yields garbage out.  Bad math is probably responsible for a smaller share of the bad decisions described in the OP than is bad risk analysis (though there's plenty of bad math to blame too).

And while there's no arguing with math, there is arguing with risk analysis, which explains the profusion of heated debates that take place in this forum around such topics as leveraged-investing-via-mortgage to a much greater extent than faulty math.

lonegun

  • 5 O'Clock Shadow
  • *
  • Posts: 7
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #8 on: January 11, 2018, 09:57:29 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

GuitarStv

  • Senior Mustachian
  • ********
  • Posts: 23128
  • Age: 42
  • Location: Toronto, Ontario, Canada
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #9 on: January 11, 2018, 10:01:02 AM »
If there is a mathematically correct way to do something, and you choose ANY other way for ANY other reason, then you are wrong.

Assuming that you're accounting for all variables, absolutely.

This is an important point though.  The optimum solution for accumulating wealth will be different than the optimal solution for retiring early.  It's not optimal for either scenario to have kids . . . but many people feel that this is essential to optimizing their happiness.  I figure we're all aiming to achieve maximized happiness in life.  Unfortunately, 'happiness' isn't always easy to measure in concrete ways . . . and many of the things that can lead to happiness conflict with one another.  'Correct' therefore depends on initial assumptions, which are prone to change.  It's hard to come up with a rock solid 'correct' solution built upon a foundation of mushy pleasure seeking.

:P

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #10 on: January 11, 2018, 10:05:01 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

its a shitty recommendation - these forums exist as was stated in the OP to challenge our current comfort zones to acheive something abnormal in society FIRE.  Using your logic if someone received a magical psychological benefit spending 10% more than they make they should do this regardless of math.

mm1970

  • Senior Mustachian
  • ********
  • Posts: 10880
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #11 on: January 11, 2018, 10:44:34 AM »
For me, it's often "don't let the perfect be the enemy of the good enough".

Yes, absolutely investing in the market and not pre-paying your low mortgage is mathematically better.

But for many people, it's not either-or.  You can do both.

In addition to that, pre-paying the mortgage is STILL better than blowing the money on eating out and crap.  So the forced "savings", while not better than investing in the market, is a middle ground.

As someone who went through the housing market crash - well, we were never underwater because we pre-paid our mortgage (despite the fact that the bottom of the market house value was $300,000 less than we paid for it).  Now, we still live in our house.  I know many people who lost their homes during that time, and sold at a loss and were underwater when the house sold.  That's a pretty big hit to your credit rating if you get a foreclosure.  Even if you aren't foreclosed on, it kind of sucks to sell the house and still owe the bank money.  For a house that you don't live in anymore.

As far as timing the market - well, I don't really do that - yet.  But as we get older and approach retirement age, I do see us moving our investments around to those that are less volatile. It's one thing to weather a massive downturn when you are 40, but another entirely at 65 with a kid in college.

nemesis

  • Bristles
  • ***
  • Posts: 259
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #12 on: January 11, 2018, 10:49:36 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.
Totally disagree with this.

Having the means and funds to pay off debt is just as fine as paying off the debt.   If you can pay off the debt anytime you want, you're not enslaved by debt. 

Your recommendation is a little too simple.

Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1795
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #13 on: January 11, 2018, 10:50:07 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

its a shitty recommendation - these forums exist as was stated in the OP to challenge our current comfort zones to acheive something abnormal in society FIRE.  Using your logic if someone received a magical psychological benefit spending 10% more than they make they should do this regardless of math.
I think you mentioned you had a baby due this year, which is great. Was your house purchased to accommodate the extra person or are you going to need more room. Were you paying an extra large mortgage payment on empty rooms until the arrival of kids?

Optimally the math says to purchase a smaller house and upgrade later when the kids arrive, a cheaper hose is always financially easier to carry the debt on than a larger house and leaves even more for investing. We're all prone to acting emotionally though.

Most homeowners have houses that are wrong size for their family needs and use justifications outside of the math. You'll hear stuff like "planning for the future"or "keeping the kids room for when they return from University". I do it too, I don't pretend otherwise.

If you have a house that has extra bedrooms (at some point), you probably did some behavioral economics yourself.

VoteCthulu

  • Bristles
  • ***
  • Posts: 409
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #14 on: January 11, 2018, 10:51:37 AM »
The primary purpose of my money is to make me happy. Sometimes serving that purpose requires sub-optimal purchases, such as when a I could eat at home for $3 but I choose to eat out with friends for $15 instead.

Some people are far happier paying off their mortgage ASAP, and as long as they understand the total cost of doing so, it can be a good decision for them.

nemesis

  • Bristles
  • ***
  • Posts: 259
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #15 on: January 11, 2018, 10:52:12 AM »
...

The bitcoin problem highlights the principle that we start making errors in rational analysis as soon as we get outside our comfort zone.  There might be a 1% chance that bitcoin is going to be worth a gazillion dollars, and a 99% chance it's worthless, and the market has potentially priced that correctly.  But because we don't "understand" that math, we dismiss it, and therefore give an irrational assessment of what could be a perfectly efficient market and price.
...
This seems completely baffling to me.  Who really understands bitcoin?  Even the bitcoin billionaires don't understand bitcoin. 

Bitcoin is by itself worthless, but there are enough people who are intrigued by the possible of "lottery win" of bitcoin to want to speculate in it (not investing).  It is worse than investing in Tulips because you at least have something physical with tulips.

Bitcoin is just speculation. It has no fundamental value except the energy used to mine it, and if people are willing to pay for that scarcity. 

Your argument about bitcoin seems bizarre and ill-placed here.

BookLoverL

  • Bristles
  • ***
  • Posts: 431
  • Location: England
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #16 on: January 11, 2018, 10:53:19 AM »
Out of the many things I learnt when I got my degree in maths, here are a couple that seem relevant:

1) Maths is 100% correct, IF AND ONLY IF the internally consistent axioms/assumptions that you started with were correct. And as discovered by Gödel, it is impossible to prove your axioms/assumptions only using the same set of axioms/assumptions.

In this case, that means that Boarder42's maths is correct, IF AND ONLY IF the efficient market hypothesis is true, a 3% 30 year mortgage is available to you, and the stock market this century behaves not too outrageously differently to the stock market last century, by always, on average, trending up. If you, beginning from some other logical source, believe these assumptions to be true for you, then B42 is right from a maths perspective for you. (I say for you because the one about 3% 30 year mortgages will vary from place to place. The other things will be equally true for everyone.)

2) You can have as much completely true maths written down as you want, but unless you know how it's relevant, it's going to be useless to you.

There were a lot of lemmas/theorems that we had to prove that I felt, afterwards, something like, "OK... but what do I do with this information?" This is because a lot of maths at the university level is written in a very abstract and general way, which means that you need a deeper understanding to be able to actually do something with it.

In this case, if you don't understand the maths and the assumptions at least well enough to feel confident in them, or if you know you OUGHT to invest but for whatever reason you haven't done it yet, or there is some other understanding or personality factor at work, you're likely going to fail at applying the maths. You may feel confident it will work for other people, but feel unable to actually apply it yourself. This might be another reason why people might choose to ignore b42.

3) Maths as practised in the science, engineering, and social sciences buildings is not really maths, not in the same way.

Now, of course, if these departments have used the logical rules of maths, their answer is correct given that their assumptions are correct. But maths is the only department where absolutely everything is done logically. Physicists make all sorts of assumptions to get their calculations actually calculable - hence the jokes about spherical chickens in a vacuum. Physics is usually looking at basic enough aspects of the universe for it not to be too much of a problem, but depending on the assumptions they've made, everything is going to be just a little bit off. The "softer" the science gets, the more of a problem this becomes. In biology they use statistics everywhere, and by the time you get to social sciences, it sometimes feels like people are pulling their assumptions out of a hat.

This last point is where behavioural economics comes in. Traditionally, economists have assumed that people are totally rational. Given most people that you meet, and the decisions you see them making, I'm sure you can agree that in the vast majority of cases, this is total nonsense. A lot of the old hypotheses are based on that assumption.

Behavioural economics drops the assumption that people are rational, which opens up the solution space significantly. And if you assume that people aren't rational, you find that different behaviours than before may occur in the markets than under the hypothesis that people are rational. And if people aren't rational, you may also find yourself recommending different things for different people, because some of them won't be rational enough to stick to your most rational plan. Some of those people will also be self-aware enough to realise they aren't rational in the moment, and arrange things so that their future self being irrational won't affect them as much. Really, then, any economist who actually wishes to understand the economy can't afford to disregard behavioural economics.



It's true, maths is maths. But, as you can see, in reality, it's often a little more complicated than that.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #17 on: January 11, 2018, 10:54:54 AM »
No, behavioral economics is the study of why human beings made decisions

Right.  In the context of this thread, however, behavioral economics is being used to justify mathematical incorrect decisions.  Math doesn't care why we get wrong answers.  The distributive property of multiplication doesn't give you a pass when your grandma dies.  The wrong answer is still wrong.

Then behavioral economics just becomes post-rationalization for our mistakes.  We like to blame human psychology for our inability to be rational.  This is a bug that need to be squashed.

Paying off debt has a magical psychological benefit that math does not show.

Anytime you find yourself arguing with math by invoking magic, you're probably making a terrible mistake.  Here's a tip for all you folks wearing healing crystals and reading tea leaves to make your investment decisions:  magic isn't real! You've been duped! 

It's hard to come up with a rock solid 'correct' solution built upon a foundation of mushy pleasure seeking.

Hard, but not impossible.  I've built a career out of putting exact numerical answers on fuzzy  questions.  You need to assign relative values to all of the things you think matter in your decision matrix, then plug and chug through the math.  We can (and should) argue about how to value different factors, but in a surprising fraction of cases it turns out to matter a whole lot less often than we expect, after we turn the crank on the math. 

In the example at hand, I agree that there is a warm fuzzy feeling from paying off your mortgage.  But you can buy a whole lot more warm fuzzies with an extra $200,000 in the bank from optimally managing your debts vs your investments.  Don't overpay for your happiness.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #18 on: January 11, 2018, 10:59:17 AM »
Behavioural economics drops the assumption that people are rational, which opens up the solution space significantly.

I feel like this is making my point for me.  Isn't this just doublespeak for "wrong answers aren't wrong anymore if you redefine what wrong means"?

As a descriptive tool for explaining WHY people make wrong decisions, behavioral economics is a useful tool.  It should never be used to justify deliberately making wrong decisions, though.

frugalnacho

  • Walrus Stache
  • *******
  • Posts: 5055
  • Age: 41
  • Location: Metro Detroit
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #19 on: January 11, 2018, 11:00:54 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

its a shitty recommendation - these forums exist as was stated in the OP to challenge our current comfort zones to acheive something abnormal in society FIRE.  Using your logic if someone received a magical psychological benefit spending 10% more than they make they should do this regardless of math.
I think you mentioned you had a baby due this year, which is great. Was your house purchased to accommodate the extra person or are you going to need more room. Were you paying an extra large mortgage payment on empty rooms until the arrival of kids?

Optimally the math says to purchase a smaller house and upgrade later when the kids arrive, a cheaper hose is always financially easier to carry the debt on than a larger house and leaves even more for investing. We're all prone to acting emotionally though.

Most homeowners have houses that are wrong size for their family needs and use justifications outside of the math. You'll hear stuff like "planning for the future"or "keeping the kids room for when they return from University". I do it too, I don't pretend otherwise.

If you have a house that has extra bedrooms (at some point), you probably did some behavioral economics yourself.

It depends on your time frame. Buying and selling homes comes with huge transaction costs, not to mention the stress of it all.  If you go from single to full family of 5 in a matter of 5 years you are probably better off over buying to start with rather than eating 6% in transaction costs each time you upgrade slightly.

soupcxan

  • Stubble
  • **
  • Posts: 181
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #20 on: January 11, 2018, 11:03:06 AM »
Boarder42 applies that analysis dogmatically with respect to pre-paying a mortgage (at least at locked-in low interest rates for an extended term).  Over a 30-year period, the market has always had greater returns than the 3% or 4% that current mortgages are at, so from boarder's perspective, it's just a math problem.  The math says invest in the market over pre-paying, so that's what you should do, no exceptions.  If you want to pre-pay your mortgage to feel more secure, etc., that's fine, but understand that you're making a bad math decision and at least know the true cost of your irrational desire for security.

It's only a math problem if you are sure the future will repeat the past. While the future is very likely to look like the past, there is a non-zero chance it doesn't. You may estimate the risk of a macro black swan type outcome at <0.1% and therefore ignore it, and someone else might put it at >1% and not ignore it, but that doesn't mean that one person is bad at math.

It will always be true that leverage increases your risk. The fact that using leverage over a 30-year period has worked in the last century does not mean that it will be risk-free in the next century.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #21 on: January 11, 2018, 11:10:02 AM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

its a shitty recommendation - these forums exist as was stated in the OP to challenge our current comfort zones to acheive something abnormal in society FIRE.  Using your logic if someone received a magical psychological benefit spending 10% more than they make they should do this regardless of math.
I think you mentioned you had a baby due this year, which is great. Was your house purchased to accommodate the extra person or are you going to need more room. Were you paying an extra large mortgage payment on empty rooms until the arrival of kids?

Optimally the math says to purchase a smaller house and upgrade later when the kids arrive, a cheaper hose is always financially easier to carry the debt on than a larger house and leaves even more for investing. We're all prone to acting emotionally though.

Most homeowners have houses that are wrong size for their family needs and use justifications outside of the math. You'll hear stuff like "planning for the future"or "keeping the kids room for when they return from University". I do it too, I don't pretend otherwise.

If you have a house that has extra bedrooms (at some point), you probably did some behavioral economics yourself.

i have a larger house than we need i bought it knowing it was larger than we need - it came at quite a discount and was purchased with a loan thats sub inflation i did the calculations to determine how it would affect my FIRE time line and chose to purchase it.  what the above poster said that i commented on was none of this they said pay off debts it feels good i dont care about math.  We wanted to be lake front - to buy a lot and build a house would have been much much more expensive than even the additional costs over many years to heat and cool a much larger space than we need. houses historically increase at inflation and my mortgage rate is at inflation for 30 years so my largest cost of ownership is just the little green soldiers tied up in the equity.  If i could have a 100% loan forever

unless you live like a monk everything you have could be attributed to behavioral economics.  i'm not saying we should ignore behavioral economics i'm saying we should still rationally look at the math and then determine if what ever you have is worth it and adds the value to your life compared to what it detracts from a financial stand point.  while i believe almost no one in the US should pay down a mortgage today that doesnt mean you cant choose to after evaluating and understanding its likely impact to your life and your time to FIRE

I've posted many times about the behaviors that eat up life energy that people choose to do like mowing their own lawn or cooking their own meals and more often than not all of these things even combined are worth less than the simple change from auto buying VTSAX instead of autopaying down principal. 

BookLoverL

  • Bristles
  • ***
  • Posts: 431
  • Location: England
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #22 on: January 11, 2018, 11:10:40 AM »
Behavioural economics drops the assumption that people are rational, which opens up the solution space significantly.

I feel like this is making my point for me.  Isn't this just doublespeak for "wrong answers aren't wrong anymore if you redefine what wrong means"?

As a descriptive tool for explaining WHY people make wrong decisions, behavioral economics is a useful tool.  It should never be used to justify deliberately making wrong decisions, though.

Of course irrational decisions are still irrational. However, given that we know people are irrational, it then becomes irrational to expect yourself or anyone else to always be rational. If somebody knows enough about their own psychology to know that they'll be overly tempted to sell at the lows if they still have debt, for instance, then expecting themselves not to because "that's the rational decision" is like an overweight person on a diet keeping a cupboard full of junk food in their kitchen (perhaps for a relative) and expecting themselves not to eat it because "that's the rational thing to do". Eventually, their temptation is going to get them, and if they'd just thrown out that junk food/made an investing decision that took into account their personality, they may have been able to avoid it. Therefore, the rational thing to do is to expect that you will be irrational in the ways that you know yourself to be irrational and to plan for that.

Another example: if someone consistently comes to leave the house and finds they've forgotten something at the last moment, and therefore frequently ends up leaving the house ten minutes later than they planned, who's more likely to get to their job on time: the person who says, "next time, I won't forget anything, and I will leave on time", or the person who says, "well, I'll probably forget something next time too, knowing me, so I'll plan to leave the house ten minutes earlier"?

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #23 on: January 11, 2018, 11:17:16 AM »
No, behavioral economics is the study of why human beings made decisions

Right.  In the context of this thread, however, behavioral economics is being used to justify mathematical incorrect decisions.  Math doesn't care why we get wrong answers.  The distributive property of multiplication doesn't give you a pass when your grandma dies.  The wrong answer is still wrong.

Then behavioral economics just becomes post-rationalization for our mistakes.  We like to blame human psychology for our inability to be rational.  This is a bug that need to be squashed.

Ah, here perhaps is the divergence in our thinking. You feel that I'm deploying behavioral economics to justify my own bad decisions. I see it as providing extremely useful information about why other people -- who I cannot control -- often make decisions that are not utility maximizing.

If I can better understand why they are making those decisions, I'm in a much better position to give advice or suggestions to get them to make decisions which are closer to optimal than the ones they are currently making than if I'm just shouting at them about how they're wrong ... and sadly also a much better position than if I'm walking them through the math on why their decisions aren't in their own long term self interest over and over again.
« Last Edit: January 11, 2018, 11:29:42 AM by maizeman »

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #24 on: January 11, 2018, 11:51:57 AM »
What is rational and irrational depends on a human value judgment, sorry people. Our rules of logic don't exist in some Platonic heaven outside of human existence and language. In fact, there is a good amount of empirical evidence that cause and effect aren't foundational characteristics of our universe, even if it often appears that way.

Whether a decision is the "right" decision depends on what you value. A good example was raised earlier.

If you accept the assumption that the ultimate and best use of your life and money is to maximize your returns, which is being argued by some of the people in this thread, then you cannot justify having a child.

Most people use money as a tool for their happiness. If happiness is the goal, the value of debt reduction is therefore subjective, and you can't say what the exact benefit is from one person to the next. You can argue that it doesn't belong in a mathematical analysis, but the ultimate goal of the mathematical analysis is to generate happiness through financial independence. If debt reduction is relevant to happiness, it follows that it is worth considering for some people.

Pretty much everything that is espoused on this website is a subjective perspective that isn't 'right' by some "objective", true nature. We just all find it appealing and value the perks of the lifestyle espoused.

thats all well and good around here we discuss how to optimize and promote the best paths forward from an optimal use of green soldiers - its easy to see i pay someone 1500 a year to mow my lawn if i cut that cost i save 1500 - its harder to grasp and understand why paying down a mortgage is a very very sub optimal decision and people fall into a few categories on this

1. ignorant and dont care to learn - frequently post poor mathematical justifications or just talk about the feeling side - this is face punch worthy
2. new and ignorant but open to understanding - could easily be lead astray by bad mathematics
3. understand all the math and how it affects them and have chosen to pay it down - happiness justification or whatever it happens to be - i have no real issue with these people except when they promote their stance to others as a good path to take.
4. understand the math and choose to not pay it down. 

obviously my biggest issues lie with number 1 and number 3 b/c it is not valuable to this forum to promote buying mcmansions or trucks or boats or 10000 dollar vacations or paying down low fixed rate mortgages etc.  Can you choose to do it and justify it internally - i guess we're now calling that  behavioral economics - yes you can.

For whatever reason - societal norms of debt being bad or maybe looking at your costs and going my mortgage is my largest cost if i get rid of that i dont need as much money and i FIRE faster.  These are the easy warm fuzzy feeling ways people approach it but people should control their behavior to the point that they at least get to number 3 and understand what in fact they are giving up to do it. 

And as i posted above if you cant control your behavior to the point of learning and understanding what is really simple math behind why keeping low fixed mortgages is better than not.  Then you may really need to evaluate if you need to Hire someone and pay them 1% of your money to manage it for you b/c that same person is likely to panic in a market downturn when FIREd.

b/c cyclestache raises a very good question. 
« Last Edit: January 11, 2018, 11:55:43 AM by boarder42 »

big_owl

  • Handlebar Stache
  • *****
  • Posts: 1051
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #25 on: January 11, 2018, 12:19:56 PM »
I don't really have a dog in this fight since I'm paying my mortgage off early AND investing at the same time so whatever, I'm happy.  Has it been figured out that although in the past the market vs. mortgage would be a better mathematical decision that this is guaranteed to be the case going forward?  I guess I need to wait and see what happens during the next recession with interest rates already at zero before I decide what I think.  That hasn't happened in modern times so I'm still skeptical at assuming the math guarantees the market will win vs. reducing debt during the next major downturn.  But again, I just do both so I'm covered either way. 


BookLoverL

  • Bristles
  • ***
  • Posts: 431
  • Location: England
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #26 on: January 11, 2018, 12:26:18 PM »
What is rational and irrational depends on a human value judgment, sorry people. Our rules of logic don't exist in some Platonic heaven outside of human existence and language. In fact, there is a good amount of empirical evidence that cause and effect aren't foundational characteristics of our universe, even if it often appears that way.

Whether a decision is the "right" decision depends on what you value. A good example was raised earlier.

If you accept the assumption that the ultimate and best use of your life and money is to maximize your returns, which is being argued by some of the people in this thread, then you cannot justify having a child.

Most people use money as a tool for their happiness. If happiness is the goal, the value of debt reduction is therefore subjective, and you can't say what the exact benefit is from one person to the next. You can argue that it doesn't belong in a mathematical analysis, but the ultimate goal of the mathematical analysis is to generate happiness through financial independence. If debt reduction is relevant to happiness, it follows that it is worth considering for some people.

Pretty much everything that is espoused on this website is a subjective perspective that isn't 'right' by some "objective", true nature. We just all find it appealing and value the perks of the lifestyle espoused.

thats all well and good around here we discuss how to optimize and promote the best paths forward from an optimal use of green soldiers - its easy to see i pay someone 1500 a year to mow my lawn if i cut that cost i save 1500 - its harder to grasp and understand why paying down a mortgage is a very very sub optimal decision and people fall into a few categories on this

1. ignorant and dont care to learn - frequently post poor mathematical justifications or just talk about the feeling side - this is face punch worthy
2. new and ignorant but open to understanding - could easily be lead astray by bad mathematics
3. understand all the math and how it affects them and have chosen to pay it down - happiness justification or whatever it happens to be - i have no real issue with these people except when they promote their stance to others as a good path to take.
4. understand the math and choose to not pay it down. 

obviously my biggest issues lie with number 1 and number 3 b/c it is not valuable to this forum to promote buying mcmansions or trucks or boats or 10000 dollar vacations or paying down low fixed rate mortgages etc.  Can you choose to do it and justify it internally - i guess we're now calling that  behavioral economics - yes you can.

For whatever reason - societal norms of debt being bad or maybe looking at your costs and going my mortgage is my largest cost if i get rid of that i dont need as much money and i FIRE faster.  These are the easy warm fuzzy feeling ways people approach it but people should control their behavior to the point that they at least get to number 3 and understand what in fact they are giving up to do it. 

And as i posted above if you cant control your behavior to the point of learning and understanding what is really simple math behind why keeping low fixed mortgages is better than not.  Then you may really need to evaluate if you need to Hire someone and pay them 1% of your money to manage it for you b/c that same person is likely to panic in a market downturn when FIREd.

b/c cyclestache raises a very good question. 

Behavioural economics isn't an excuse to do what you wish, though. It's about recognising that people, and therefore you, and me, and MMM, and Warren Buffett, and Albert Einstein - everyone - is irrational in some way or another. It'll be different for each person, of course.

People can be good enough at maths to understand the argument for keeping a mortgage, and indeed good enough at investing to make it not worth it for them to pay one percent fees, without necessarily having a temperament that lets them make good decisions while in debt. If the debt causes them to lose sleep, and the gradual sleep deprivation causes them to make 100 tiny decisions each day that were just that little bit worse than otherwise, and they end up impulse-purchasing, or eating a whole box of donuts, or yelling at their family, and then they get so stressed that they draw money out of their investments and use it to pay down their mortgage anyway just to get rid of the stress, would using their mortgage as leverage be right for them? The interest rates haven't changed on either, but the situation looks quite different.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #27 on: January 11, 2018, 12:28:44 PM »
I don't really have a dog in this fight since I'm paying my mortgage off early AND investing at the same time so whatever, I'm happy.  Has it been figured out that although in the past the market vs. mortgage would be a better mathematical decision that this is guaranteed to be the case going forward?  I guess I need to wait and see what happens during the next recession with interest rates already at zero before I decide what I think.  That hasn't happened in modern times so I'm still skeptical at assuming the math guarantees the market will win vs. reducing debt during the next major downturn.  But again, I just do both so I'm covered either way.

1st just about everyone here who is paying down their mortgage is investing as well - b/c if they arent they are greatly losing tax advantaged space that should be utilized fully prior to considering a paydown. so you do have a dog in the fight you're on the paydown side.

2nd no we cant guarantee any return but we assume returns will be similar to history.  We use the same assumptions that most here using the 4% SWR in retirement will be using so if you dont believe those assumptions to be true you'll likely have to work decades longer. 

3rd you cant time the market and thats what you're trying to indicate with your last comment

stastically you are extremely unlikely to hit the perfect paydown window in which paying down your mortgage then investing your full payment would have beaten investing in the market and letting your mortgage go to term.  its worse than betting on green on a roulette wheel. 

So if we cant time the market what do we do? we choose the most statistically(mathmatical) likely path to help us atain our goal (FIRE here)  could you get lucky and hit a 2000 to 2010 paydown window and come out ahead yep - odds you hit that slim to none.

big_owl

  • Handlebar Stache
  • *****
  • Posts: 1051
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #28 on: January 11, 2018, 12:41:09 PM »
I don't really have a dog in this fight since I'm paying my mortgage off early AND investing at the same time so whatever, I'm happy.  Has it been figured out that although in the past the market vs. mortgage would be a better mathematical decision that this is guaranteed to be the case going forward?  I guess I need to wait and see what happens during the next recession with interest rates already at zero before I decide what I think.  That hasn't happened in modern times so I'm still skeptical at assuming the math guarantees the market will win vs. reducing debt during the next major downturn.  But again, I just do both so I'm covered either way.

1st just about everyone here who is paying down their mortgage is investing as well - b/c if they arent they are greatly losing tax advantaged space that should be utilized fully prior to considering a paydown. so you do have a dog in the fight you're on the paydown side.

2nd no we cant guarantee any return but we assume returns will be similar to history.  We use the same assumptions that most here using the 4% SWR in retirement will be using so if you dont believe those assumptions to be true you'll likely have to work decades longer. 

3rd you cant time the market and thats what you're trying to indicate with your last comment

stastically you are extremely unlikely to hit the perfect paydown window in which paying down your mortgage then investing your full payment would have beaten investing in the market and letting your mortgage go to term.  its worse than betting on green on a roulette wheel. 

So if we cant time the market what do we do? we choose the most statistically(mathmatical) likely path to help us atain our goal (FIRE here)  could you get lucky and hit a 2000 to 2010 paydown window and come out ahead yep - odds you hit that slim to none.

I guess I was trying to say that I'm going to continue to invest significantly and pay down mortgage significantly no matter what the argument is so that's why I don't really care about it.  There's going to be opportunity cost I'm missing out on no matter which way it goes since I'm doing both, but I've found the best way for me to feel better is to significantly grow active income so that we can do both those things simultaneously and still afford to do what we want in life and feel good about it.

We'll see what happens after the next big downturn, nobody knows what will happen.  Maybe it'll be long term debt hangover or maybe the fed will price target and inflate away. 

Maenad

  • Pencil Stache
  • ****
  • Posts: 643
  • Location: Minneapolis 'burbs
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #29 on: January 11, 2018, 12:43:20 PM »
I find it irritating that people are confounding optimal/suboptimal with right/wrong.

Yes, there is probably one financially optimal answer for each question, but that doesn't make the others "wrong", just "suboptimal". A 60/40 stock/bond split may not have as much longevity as 85/15 if you're planning for a 60-year payout, and may therefore require a lower WR in order to make it to 95% confidence, but if someone is willing to work longer for that lower WR, and the 60/40 lets them sleep at night and keeps them from panicking in a down market, that's a hell of a lot better than insisting on 85/15 and that they should magically become more "rational" and risk-tolerant.

Encouraging people to be more rational and data-driven is awesome. Beating them about the head and shoulders with the math is jerkish. And I say this as an engineer who spends a lot of time in risk management.

Also, re: market efficiency. The markets may be overall efficient, but sectors can and do go into bubbles on a fairly regular basis. Efficient Market Theory and the existence of bubbles are not mutually exclusive. Read up on The Growth Stock/New Issue Craze (also known as the 'tronics boom) that started in the late 1950s, the Conglomerate Boom in the mid-60s, "performance investing" of concept-stocks of the late 60s, the Nifty Fifty in the 70s, the New Issues and biotech stocks in the 80s, plus of course the internet bubble in the late 90s. You'll see all of the same signs as we're seeing in crypto today. Forty-third verse, same as the first.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #30 on: January 11, 2018, 12:50:40 PM »
@BookLoverL  i did not say everyone who chooses to pay down their mortgage should probably enlist a financial advisor - i said those who do so with no regard to the math b/c of those feelings you listed and refuse to learn or understand the math b/c of feelings likely have a higher predisposition to also freaking out in market down turns.

Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1795
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #31 on: January 11, 2018, 01:12:17 PM »
Paying off debt has a magical psychological benefit that math does not show. It lifts a heavy burden from your mind when complete. I almost always would recommend paying off all debt prior to investing even if it bad math.

its a shitty recommendation - these forums exist as was stated in the OP to challenge our current comfort zones to acheive something abnormal in society FIRE.  Using your logic if someone received a magical psychological benefit spending 10% more than they make they should do this regardless of math.
I think you mentioned you had a baby due this year, which is great. Was your house purchased to accommodate the extra person or are you going to need more room. Were you paying an extra large mortgage payment on empty rooms until the arrival of kids?

Optimally the math says to purchase a smaller house and upgrade later when the kids arrive, a cheaper hose is always financially easier to carry the debt on than a larger house and leaves even more for investing. We're all prone to acting emotionally though.

Most homeowners have houses that are wrong size for their family needs and use justifications outside of the math. You'll hear stuff like "planning for the future"or "keeping the kids room for when they return from University". I do it too, I don't pretend otherwise.

If you have a house that has extra bedrooms (at some point), you probably did some behavioral economics yourself.

i have a larger house than we need i bought it knowing it was larger than we need - it came at quite a discount and was purchased with a loan thats sub inflation i did the calculations to determine how it would affect my FIRE time line and chose to purchase it.  what the above poster said that i commented on was none of this they said pay off debts it feels good i dont care about math.  We wanted to be lake front - to buy a lot and build a house would have been much much more expensive than even the additional costs over many years to heat and cool a much larger space than we need. houses historically increase at inflation and my mortgage rate is at inflation for 30 years so my largest cost of ownership is just the little green soldiers tied up in the equity.  If i could have a 100% loan forever

unless you live like a monk everything you have could be attributed to behavioral economics.  i'm not saying we should ignore behavioral economics i'm saying we should still rationally look at the math and then determine if what ever you have is worth it and adds the value to your life compared to what it detracts from a financial stand point.  while i believe almost no one in the US should pay down a mortgage today that doesnt mean you cant choose to after evaluating and understanding its likely impact to your life and your time to FIRE

I've posted many times about the behaviors that eat up life energy that people choose to do like mowing their own lawn or cooking their own meals and more often than not all of these things even combined are worth less than the simple change from auto buying VTSAX instead of autopaying down principal.
I agree about using logic to maximize your choices. 

For the record, you made a math mistake. House prices do not track inflation, they track wage growth. Inflation also tracks wage growth (in general), so why mention it? Where its important though is if you're buying a house in a region with a depressed, stagnant or booming economy. House prices will not track national inflation rates, they will track local wage growth (or shrink in areas with massive job losses - coal country perhaps? the small town I was born in?). Those same regions that see house prices decrease may see overall inflation, as fuel, food and the rest continue to rise.

There is a famous example on this board of someone who recognized this - ARS. He recognized that the wages in Las Vegas did not decrease during the housing crisis while the house prices did. Over 9 years he bought a bunch of house prices that had dipped from the predicted prices, based on wages, and made a small fortune. Contrast that with some regions where hosing and wages decreased, fortunes weren't made there. I'm over simplifying, its hard for people to get when first introduced (otherwise they wouldn't have made the error to begin with).

Perhaps you should revisit your math though and plugin the forecast wage growth for your region instead of inflation, would it really matter though? You might notice though, while inflation tends to be predicted for the long term, wage growth isn't. If it was easy to predict wage growth, we all would have bought houses in Silicon valley 30 years ago.

For every graph you can produce that shows housing tracking inflation, you can find one that shows housing tracking wage growth (with greater accuracy). Compare for yourself, all three look similiar but wages explain it on the small scale (the most dramatic is small towns that lose their main employer).

P.S. I enjoy your comments, you're fun to respond to (I like that you don't hold back).

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #32 on: January 11, 2018, 01:24:56 PM »
you're right they track wage growth not inflation. 


i really like this topic - the more i think about it if a person cannot take the time to put emotion aside and calc something and then determine if its worth it to them i think this is a real problem and it could show its ugly head in a downturn in FIRE in a very bad way. 

Scortius

  • Bristles
  • ***
  • Posts: 475
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #33 on: January 11, 2018, 01:31:52 PM »
Encouraging people to be more rational and data-driven is awesome. Beating them about the head and shoulders with the math is jerkish. And I say this as an engineer who spends a lot of time in risk management.

I actually feel like this is where a big disconnect is coming from. Boarder doesn't argue that you should maximize expectation in a vacuum. He certainly doesn't "beat them over the head in a jerkish manner", yet when he does come to offer his analysis people who disagree with him often act offended and insulted.

Here's what I see happening. I see people come into threads or start new threads where the topic will involve paying off their mortgage early for a number of reasons, very often including desires to 1) reduce risk of bankruptcy when losing their job, 2) pay off the mortgage quickly so they can invest the savings into the market and make more money once it's paid off, or in general 3) lower the time it takes for them to achieve financial independence. Others incorrectly assume you have to increase your FI stach by 25x your mortgage (you don't, you just need to save an additional amount equal to your outstanding pay-down balance). Others complain that they shouldn't be 'lectured' because they are holding higher rate loans, ARMs, or are not in the US where 30 year fixed are available (and Boarder and others are always quick to point out that this analysis is ONLY for people with 30 year fixed at ~4% and below). Others even come in here and talk about paying off their mortgage before even maxing their tax-advantaged space.

The point is that for people who emotionally want to prioritize any of these goals, it is highly advantageous to not pay off the mortgage. Boarder isn't arguing to these people that they have to maximize their expectation regardless of risk.  He is arguing that if they are being truthful to themselves about their own peace-of-mind goals, then they are going about it the wrong way. His advice generally benefits many people who are risk-adverse but don't understand that they are actually choosing a path with a higher risk of ruin. Hell, for those with fixed 30 years at 3.5 or lower, it's not unreasonable to expect that in a few years you'll be able to buy bonds that will outperform your mortgage. That's as much of a risk-adverse win as you could ever find.

So no, I reject that those in the "don't pay" camp are arguing from a robotic risk-agnostic perspective. Quite the opposite, people who want to lower their risk are often very unaware that the choices they are making are both lowering their expectation and increasing their risk of ruin. This is why so many of us continue to push the debate.

neil

  • Stubble
  • **
  • Posts: 215
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #34 on: January 11, 2018, 01:39:23 PM »
Are people really doing the math?  I think we ignore boatloads of available data because it is information overload, and justify it for various reasons.  And to invoke behavioral into that context, we generally don't revisit assumptions unless those assumptions directly and provably cause us harm.  It's easy to forget the "math" involved here is statistics and not algebra and it is heavily data and variable dependent.  Human beings are deciding on the variables and that makes the conclusions biased in some percentage based on behavioral factors. 

The reason I can FIRE is because I'm willing to execute on backup plans, and accept personal circumstances might fuck me regardless.  Not because a FIRE tool said 4% is ok.  Economic models are nice and pretty... and largely piles of shit, as proven over and over by massive historical failures of anyone trying to predict things with them.

I don't think the tools are shit but there's some odd kind of reverence for 150 years of sequential stock returns and using that to make predictions on 50 year retirement schemes.  It's the best we can do, but humans as a collective have no proven track record for getting this right.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #35 on: January 11, 2018, 01:40:42 PM »
well said @Scortius and that specific behavoir that many take to getting up in arms and refusing to understand - i think is what the OP of this thread was trying to get at - and i would think that if you're going to act irrationally about a large financial decision even when presented with data that your path actually harms your goals - then this is something we should be discussing on a much deeper level here.

The issue at hand is people act irrationally even when they think they are being rational and Scortius points out 3 very common reasons brought up - So if they cannot accept the data presented them and change their path on the mortgage - how can one reasonably expect that they personally will hold their stocks to their intended AA during a down turn in FIRE.  I would think they have a much higher risk of actually failing at FIRE if a downturn happened due to their past history of being presented data and for the most part ignoring it.  so when someone here says just hold to your AA they say no the markets tanking i have to get out.

Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1795
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #36 on: January 11, 2018, 01:41:39 PM »
you're right they track wage growth not inflation. 


i really like this topic - the more i think about it if a person cannot take the time to put emotion aside and calc something and then determine if its worth it to them i think this is a real problem and it could show its ugly head in a downturn in FIRE in a very bad way.
Wow, I was really expecting you to disagree, most people do out of instinct. You can actually apply this to the positive, I might buy a house in the city Amazon chooses for its second head office for example.  Ultimately I'm a little bit lazy ( or busy with other more interesting endeavours) I tend towards the easy route of buying ETF.

You seemed to accept it too quickly, perhaps you have some prime real estate on Vulcan?

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #37 on: January 11, 2018, 01:46:04 PM »
you're right they track wage growth not inflation. 


i really like this topic - the more i think about it if a person cannot take the time to put emotion aside and calc something and then determine if its worth it to them i think this is a real problem and it could show its ugly head in a downturn in FIRE in a very bad way.
Wow, I was really expecting you to disagree, most people do out of instinct. You can actually apply this to the positive, I might buy a house in the city Amazon chooses for its second head office for example.  Ultimately I'm a little bit lazy ( or busy with other more interesting endeavours) I tend towards the easy route of buying ETF.

You seemed to accept it too quickly, perhaps you have some prime real estate on Vulcan?

its been discussed at length here before its easy/common to say inflation but wage growth is really what drives it and i've seen the data.  i'm currently making a killing on my lake front mcmansion easily sitting over 100% ROI on what we've got "invested" in it.

Stachless

  • Stubble
  • **
  • Posts: 128
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #38 on: January 11, 2018, 10:01:27 PM »
I do believe Pete (the dude who's blog we are all on right now) paid his mortgage off early.  Maybe even twice.

Boarder42, I agree with your math and appreciate what you do here fighting the good fight.  I was surprised at first when so many were offended by your comments in this 'cult' of 'facepunchers', but after really thinking about how hard so many of these good folks worked and all the sacrifices they made to do so, can better understand why they are so emotionally invested (or not invested!) in paying off the mortgage.

I see this topic as being a lot like sex.  Just because doing it one way may be 'optimal' doesn't mean the other ways of doing it aren't great!

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #39 on: January 12, 2018, 04:25:53 AM »
I do believe Pete (the dude who's blog we are all on right now) paid his mortgage off early.  Maybe even twice.

Boarder42, I agree with your math and appreciate what you do here fighting the good fight.  I was surprised at first when so many were offended by your comments in this 'cult' of 'facepunchers', but after really thinking about how hard so many of these good folks worked and all the sacrifices they made to do so, can better understand why they are so emotionally invested (or not invested!) in paying off the mortgage.

I see this topic as being a lot like sex.  Just because doing it one way may be 'optimal' doesn't mean the other ways of doing it aren't great!

Pete had much higher mortgage rates that wouldn't apply to today's scenario.

markbike528CBX

  • Handlebar Stache
  • *****
  • Posts: 1899
  • Location: the Everbrown part of the Evergreen State (WA)
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #40 on: January 12, 2018, 06:35:02 AM »
Great discussion.   While I'm one of the suboptimal, emotional mortgage killers ( just did it), I spent most of my life saving, then investing when I realized that simply saving wasn't optimal. 
 Since I now have "enough", a comfortable 4% Fire stache, plus 20% more, I don't have to maximize utility to increase my happiness.

My advantage is that I didn't have the options of mortgage pay down vs investing, since I had no house (single, didn't need/want one).

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #41 on: January 12, 2018, 07:25:17 AM »
First, thank you for all the thoughtful responses.  I spent some time with this, so I appreciate the insight.  A couple responses:

...

The bitcoin problem highlights the principle that we start making errors in rational analysis as soon as we get outside our comfort zone.  There might be a 1% chance that bitcoin is going to be worth a gazillion dollars, and a 99% chance it's worthless, and the market has potentially priced that correctly.  But because we don't "understand" that math, we dismiss it, and therefore give an irrational assessment of what could be a perfectly efficient market and price.
...
This seems completely baffling to me.  Who really understands bitcoin?  Even the bitcoin billionaires don't understand bitcoin. 

Bitcoin is by itself worthless, but there are enough people who are intrigued by the possible of "lottery win" of bitcoin to want to speculate in it (not investing).  It is worse than investing in Tulips because you at least have something physical with tulips.

Bitcoin is just speculation. It has no fundamental value except the energy used to mine it, and if people are willing to pay for that scarcity. 

Your argument about bitcoin seems bizarre and ill-placed here.

This is a good example of believing you know more than the market and why it's so hard to internalize the correct rational analysis.  You can't simultaneously hold the view that you don't understand the market and yet you know that it's worthless.  Bitcoin may well be worthless.  But there's enough money being put into estimates of its potential worth to create a sufficient market, and so you have to ask what informational advantage do you have that allows you to know more than the market.  If something is "clear" to you (or here, clear that it's unclear), then why is it not clear to the $224 billion market for bitcoin?  We understand that analysis with respect to Ford when we invest in index funds.  I may see a new Ford and think wow that looks great, maybe Ford will go up, but I don't invest extra in Ford because I realize that the market for Ford is large enough that my information (the new model looks good to some people) has already been priced in.  There may be a 99% chance that bitcoin is worthless and a 1% chance that it's worth a gazillion dollars, and the market can therefore be rational even though it seems crazy.  The "rational" analysis for considering any market that is sufficiently developed is what informational advantage do I have.  We get it for the index funds, but it's tougher to apply to a new potential market.

"Behavioral Economics" is just code for "bad math".

The beauty of math is that it is correct.  There is no arguing with it.  Your feelings don't matter.  If there is a mathematically correct way to do something, and you choose ANY other way for ANY other reason, then you are wrong. 

Which is fine.  Humans make deliberately wrong decisions all the time.  But we don't need to pretend that our wrong decisions are actually right decisions just to make ourselves feel better.  We don't need to resort euphemisms like behavioral economics.  Just admit you've deliberately made suboptimal choices because you're bad at math, and we can all move on.

@maizeman  and @BookLoverL provided excellent responses to this, but I wanted to emphasize the point because Sol's post highlights a common misunderstanding of behavioral economics that is central to understanding the point I'm trying to make above.

Behavioral economics is not about bad math.  It's about people making irrational decisions that they believe to be rational at the time, in consistent and predictable ways.

If I asked you if you wanted to go buy some cookies right now, there's a good chance you'd say no.  Don't need the extra calories, not that healthy, etc.  But if I put a bag of cookies in front of you, you're statistically more likely to take one.  If I then put you under stress, you're even more likely to take one.  If I ask you to work on a hard math question, you're statistically more likely to take a cookie.

Why?  The analysis from your perspective never changed.  You ability to do the cookie math (is short-term enjoyment worth long-term unhealthiness) never changed.  But you--and everyone else--are statistically more likely to take a cookie when I change fact patterns that should have no rational impact on the cookie math.

Even more interesting is what happens when you ask people what they think will happen?  Do you want a cookie?  No.  Would that change if I put a bag in front of you?  No.  What if I asked you a hard math question while the cookie is in front of you?  No, I don't want a cookie.  But statistically, you are more likely to take a cookie in those situations.

That's the point.  It's not a question of being able to do rational analysis.   It's whether people are going to make rational decisions when we know statistically that they won't, even though they believe they will.  That's why I express skepticism about the general forum's ability to hold stocks during a crash.  We are all trying to internalize the rational points that there's always a risk of a crash and that in the long-term the market goes up.  But, statistically, most people buy high and sell low, even when they believe they won't.

So the test is to figure out the extent to which we can so internalize the "math is math" principles that we continue to make rational decisions in times of crisis--in short, that we can become less human.  Or alternatively, the extent to which we need to take steps to protect ourselves (investment policy statement, automatic investing, low information diet, more conservative portfolio) from the statistically likely irrational behavior.
« Last Edit: January 12, 2018, 07:27:33 AM by Cycling Stache »

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #42 on: January 12, 2018, 07:40:08 AM »
so thats all well and good you've isolated a problem that may effect some people.  so what do we do about it and how do we determine who would be must suceptible to selling in a down turn.  just saying humans act irrationally doesnt really benefit readers and posters, we need to understand how why and when and how to mitigate it.

as i've indicated a couple times i think people who make the emotional decision of paying off their house are more likely to sell in a down turn - i also think this decreases alot if they take the time and understand the math and just determine they'd rather have it paid off for what could be a number reasons. 

I also think this community will help people not sell and be more celebratory on the way down for those in accumulation.  So the real risk lies for those who are retired. 

So IMO those at highest risk for acting irrationally in a down turn are those who are RE and chose to not understand the math enough to know holding on to their stocks and rebalancing creates the best outcome. And those people are likely the same ones who paid down there mortgage and didnt care to learn about the math. - and who likely would actually benefit from a good financial advisor that will put their money in low cost index funds and help them ride out the market dips by keeping them to their plan. 

I'm very excited for a market crash b/c i want to see how i will personally react to it, since i havent been thru one since joining up here and getting a real understanding for how things have worked in the past.  I'm pretty confident as analytical as i am it wont be hard to hold on and enjoy the ride down. 

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #43 on: January 12, 2018, 07:46:33 AM »
You ability to do the cookie math (is short-term enjoyment worth long-term unhealthiness) never changed.

The "cookie math" is not math, it is cost-benefit analysis.  There is no math involved, and therefore no unassailably-correct answer yielded by math.  And part of the reason that there is so much pushback in the threads on leveraged-investing-via-mortgage and similar topics is the tendency by some posters to conflate non-math with math (although, as I said above, bad math is also a big part of the reason, but that tends to get corrected quickly enough, because there is no arguing with math).  We can (and should) identify poor cost-benefit analysis, or poor risk analysis, without invoking math to defend non-math.  It's harder to defend those aspects of our positions that do not rely on immutable laws of the universe, but so be it.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #44 on: January 12, 2018, 08:14:27 AM »
Behavioral economics is not about bad math.  It's about people making irrational decisions that they believe to be rational at the time, in consistent and predictable ways.

I think these two sentences contradict each other.

Behavioral economics describes why people make mathematically incorrect decisions.  It's all about bad math.

Using it to understand the motivations behind bad math is fine.  Using it to justify or support deliberately making mathematically bad decisions is dumb.  We should identify our flaws in order to correct them, not celebrate and reinforce them.

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #45 on: January 12, 2018, 08:38:55 AM »
Behavioral economics is not about bad math.  It's about people making irrational decisions that they believe to be rational at the time, in consistent and predictable ways.

I think these two sentences contradict each other.

Behavioral economics describes why people make mathematically incorrect decisions.  It's all about bad math.

Using it to understand the motivations behind bad math is fine.  Using it to justify or support deliberately making mathematically bad decisions is dumb.  We should identify our flaws in order to correct them, not celebrate and reinforce them.

correct and for those who cannot come around to the math part of this simple formula and still choose to pay it down i think they are at real risk for other irrational decisions like selling in a down market.

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #46 on: January 12, 2018, 09:12:09 AM »
Using it to justify or support deliberately making mathematically bad decisions is dumb.  We should identify our flaws in order to correct them, not celebrate and reinforce them.

I agree that we should not be celebrating bad decisions.

But I also believe that we should be skeptical about your ability to act rationally, notwithstanding your belief that you will do so.  The key is that the fact that you think you will act rationally makes you no different than anyone else.  Maybe you will and maybe not, but statistically you won't, at least in certain situations. 
« Last Edit: January 12, 2018, 09:15:32 AM by Cycling Stache »

boarder42

  • Walrus Stache
  • *******
  • Posts: 9332
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #47 on: January 12, 2018, 09:19:31 AM »
Using it to justify or support deliberately making mathematically bad decisions is dumb.  We should identify our flaws in order to correct them, not celebrate and reinforce them.

I agree that we should not be celebrating bad decisions.

But I also believe that we should be skeptical about your ability to act rationally, notwithstanding your belief that you will do so.  The key is that the fact that you think you will act rationally makes you no different than anyone else.  Maybe you will and maybe not, but statistically you won't, at least in certain situations.

so whats your goal with this thread - to just point it out? or to try to determine who is more susceptible? 

cool you pointed out a problem with how people act typically - we are in a subset of society that intentionally acts differently in many ways - likely making us less likely to make the poor decision.  but some of the actions of people here probably put them in a camp that make them more likely.  i dont think people who choose to have a lawn service or larger house fit into the camp of making an irrational decision based on math as much as people who choose to pay a mortgage down b/c there is no change to your lifestyle to simply fund a taxable account over a mortgage prepayment.  just like shifting money into bonds when the market starts crashing doesnt appear to affect your lifestyle on the surface and feels safer.

Seradoc

  • 5 O'Clock Shadow
  • *
  • Posts: 44
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #48 on: January 12, 2018, 09:39:00 AM »
as i've indicated a couple times i think people who make the emotional decision of paying off their house are more likely to sell in a down turn - i also think this decreases alot if they take the time and understand the math and just determine they'd rather have it paid off for what could be a number reasons. 

I disagree that people who realize gains during an appreciating market and apply them towards a mortgage are likely the same as those who do not realize their gains and then cash out of the market immediately after a market crash.  They may disagree with and feel partonized by this assessment of their personal maturity.

I do think that this group is timing a market correction though, and that this decision is likely based off of a, likely emotional, "gut check" risk assessment.  Humans are terrible at assessing risk.

mm1970

  • Senior Mustachian
  • ********
  • Posts: 10880
Re: Towards a Unifying Theory of Math is Math and Behavioral Economics
« Reply #49 on: January 12, 2018, 09:49:34 AM »
Are they bad decisions or sub-optimal?  When do you cross over?

Was buying too big a house no matter what other great deals were had, if you don't need the space now and just have to pay more for it, or to heat and cool it, even if you can mitigatge that (for boarder42) a bad decision, or sub-optimal?  Do we know that now, or will we have to wait 5-10 years to see how the housing market is doing, to see if he's still living in it, etc?