Comment by cylinder | Hacker News Reader

Comment by cylinder | Hacker News Reader

Parent Full thread

Looking at principal balances is all wrong. GDP means nothing in this instance either.

What matters, and what lenders look at, is debt serviceability. Lenders don't want principal paid back, ideally you just keep paying interest for life.

Let's say median household:

Income: $4000/mo

Rent: $1500/mo

Other necessities: $2000/mo

Let's say $500/mo "disposable." It's from this that the interest needs to be paid to service the debt on that $12,600 (actually, that's per person; there are 2.53 people in the median household, so $32k debt). What's the average interest rate? Maybe 10%? That's $266/month just for interest.

Now, economy goes into recession, consumer loses income and will first stop paying unsecured debt (credit cards, student loans), then auto loan until repo, then finally mortgage. And keep in mind median American doesn't have a lot of liquid savings to cushion. If there's anything, it's in 401k or home equity - and now is not a good time to sell. But many are forced to, further increasing supply when buyers are scared off, so prices keep declining.

Everything starts to be sold and assets get repriced. More seized cars at auction. More foreclosed houses. Banks restrict credit. Everything spirals down, quickly. Cashed up HNW investors willing to take the risk buy up firesale assets with cash and hold them. Maybe prices recover eventually, but now more assets are held by fewer people, and you've moved further into inequality.

The US is very much a credit driven economy. That's why the credit crunch of 2007 had such a severe impact.

This just happened a decade ago, and it seems people have already forgotten what it's like.

> What matters, and what lenders look at, is debt serviceability.

Quite true. But it also implies something else: Because the disparity of income levels in the US is so broad, just talking about the problem in terms of national totals and averages is useless. Show me the problem divided up by quantiles of the population, maybe even log-scaled as you get closer to the top, and then I might believe that there's a problem.

Don't look at quantiles, look at people over time. Quantiles conflate poor working families with middle-class college students.

A large amount of individual debt is wholly optional. What people ignore and therefor schools do not teach to be wary is that marketing is very well developed and convince otherwise rational people to make an irrational decision which has them take on more debt.

from buying too much house or car to over buying an education that cannot be used where the person is or in a field that cannot withstand the costs. all of these are marketed products. all three have vary optional levels of how much they cost.

then add in all the monthly bills and this is where many people miscalculate and it snow balls. from cell to internet to even television. you can quickly end up paying a significant amount of your income that has no tangible return.

My lesson in financial responsibility was to have to live on 8000--10000$/year for three years. Coming from a solidly middle class income, it was a real shock. I am very credit averse, now, since even a modest level of credit sunk as fixed cost with no benefit.

Buy everything with credit card. Establish your credit rating in case you need to borrow, and take advantage of whatever reward program fits you best. Always pay of the entirety of your balance every month.

> convince otherwise rational people to make an irrational decision

This is a myth that economics will not let go of. Almost all human decisions are influenced by emotions. Here's two examples:

[A] Would you rather receive $49 or have a 50% of receiving $100? - The rational decision would be to choose the 50% chance since your expected value would be $50. Most people go for the $49, humans are irrationally risk-averse in terms of gains.

[B] 1)You have $1,000 and you must pick one of the following choices: Choice A: You have a 50% chance of gaining $1,000, and a 50% chance of gaining $0. Choice B: You have a 100% chance of gaining $500.

2)You have $2,000 and you must pick one of the following choices: Choice A: You have a 50% chance of losing $1,000, and 50% of losing $0. Choice B: You have a 100% chance of losing $500.

Time after time, when presented with this problem, people chose B for question 1 and A for question 2. A rational person would choose either A or B for both questions. The expected value is the same for both questions, but it shows that people are risk averse in gains and risk seeking in mitigating losses.

Couple this with the number of people who actually understand the concept of compounded interest rates and yikes.

Just nitpicking, but people are not otherwise rational. Study after study has shown that emotions are involved in nearly every,if not all, decisions that we make.

Reducing variance is worth real money to people, since the marginal utility of the first dollar is much higher than the 1000th.

Adding to that - it's the flaw of averages at work in many of those examples. The listed "expected value" in some of those examples could never actually be a real outcome of the examples - it is completely rational to opt to take a 500 loss rather than risk a 1000 loss if you have, say, only 750 to your name and cannot be overdrawn (see debtors prison or whatever incentive one would like to imagine). The expected value of the $500 loss only happens in one option. In the $0 or $1000 loss, if you had repeated draws over time it would indeed trend towards an average of a $500 loss but in a single draw situation, you can never achieve the average so you absolutely must view it as different from the flat $500 loss and compare that to your imagined circumstances in this thought exercise.

Yes. There are many cases where a lower expected value is a rational trade off for reduced variance.

I completely agree. A bird in hand is worth 2 in the bush and all that. However, even today economic theories do not take this into account and assume rationality === choosing the highest expected payoff. I.E. we're all statistical savants

Yep, your main point that economics' rational man is a complete fiction is absolutely correct.

He's a convenient fiction though. Since to homo oeconomicus a certain expense of USD 500 and the 50 % chance of a USD 1000 expense are the same, expect him to be rolled out to sell you high-deductible health insurance and health savings accounts. H. oeconomicus is the model human, consequently he isn't cashflow constrained, and the fact that HSAs double as tax shelters for the wealthy doesn't concern the Ideal Man either.

University economics is a religion, not a science, it offers up myths instead of testable predictions on observable reality. High finance is a different deal, but they surely don't build their models on university textbooks.

> University economics is a religion, not a science

That's a bit harsh. Is every field that employs simplified approximations of complex systems a religion as well? With that definition even physics is out and that's one of the more concrete sciences.

At times it does appear to be a religion because of the cult of personality that proliferates (remember the Greenspan years?) and, most significantly, the blind faith and slavish devotion to free markets being a solution to everything.

What I find most off-putting are those who advocate for wholesale changes in our society on little more than hypothesis and faith instead of running experiments starting at the smallest reasonable scale and then successively working your way up from that.

It should at least pretend to try to applying basic scientific methodologies to its practice.Psychology has evolved to being a rigorous, evidence based social science. That psychology is not one of the foundations of studying economics is also problematic too.

If it does not admit that it's simplified assumptions are simplified assumptions and seek to improve them, then yes. Which does not apply to physics.

Economics has a weird tendency to invoke "outside shocks" to explain why their models fail, and seems to have difficulty identifying those shocks.

For a real world example, consider government bonds which have relatively low returns but are almost guaranteed.

You answer to question A is true only if this is a repeatable event. If you have only a one-time decision to make, the logical answer is to take the money and run.

You are defining rational behavior as ignoring variance and accepting expectation as a gauranteed outcome for a trial.

I don't want to be a jerk, but you are wrong to a phenomenal degree.

You're not being a jerk. I should've put quotes around rational to highlight that this was the common thinking at the time in the field. That was just a extremely example of Prospect Theory. Before this was presented as a better model to understand human behavior, the prevailing theory didn't take into account that humans are even risk averse with gains and more likely to take risks with losses. My main point was that it took psychologists to present a model of man that resembled reality

Cumulative Prospect Theory takes this into account and states that people tend to overweight extreme, but unlikely events, but underweight "average" events, but I digress

I was just trying to show the base irrationality of the human mind. Here's another example. Consider you have two offers for the New York Times:

[A] $59 – Internet Only Subscription $125 – Internet and Print Subscription

[B] $59 – Internet Only Subscription $125 – Print Only Subscription $125 – Internet and Print Subscription

Both offers are the same, with the exception of including the “print only” subscription in Offer A. Despite the fact that not a single person chose that unattractive offer, its impact was dramatic – In a study, 62% more subjects chose the combined print and Internet offer, and predicted revenue jumped 43%. The print-only offer was the decoy, and served to make the combined offer look like a better value. This is called decoy marketing and is only one of an arsenal of tools that is used to take advantage of the irrational mind. You'll see this everywhere so it pays to identify it.

Your argument is nonsensical. Your scenarios have differences in risk. Transferring risk from one party to another is the entire basis of our financial system. To assume risk is irrelevant in step one is to fundamentally misunderstand how economic decision making works.

You are missing the greater point. I agree with you. Before Prospect Theory, economists did assume risk was irrelevant and that one would choose either A or B for both. That even a layman would find it nonsensical is exactly my point.

I should've put rational in quotes to show sarcasm or followed rational with "defined by economists at that time"

> That even a layman would find it nonsensical is exactly my point.

I have a degree in economics and both my parents are professors of economics, so I have some perspective on this. Economics as a discipline has never really been confused about risk in the way you describe.

Let's look at it another way and the so-called "sunk cost" fallacy. I have somewhat expensive tickets to an NBA game. On the night of the game, I'm sick and there's a blizzard. If I paid for those tickets, I'm damn well going to that game. If they were given to me, I'm staying in. Study after study shows that most people also would act in this matter. That this type of behavior was considered irrational made me think that a lot of economists in the past didn't try to understand actual human behavior. BTW - this is not comparable to an investment situation

I have a degree in economics too and my main complaint is that I wish those in the field treated it with more rigor, more like a scientific or engineering field... explicitly state the simplified assumptions and limits on the models. And don't proclaim that your model is correct without experimenting. It's not as much a problem as it was in the past, but people like Hayek advocated for whole changes in society based upon ideology and philosophy without any experimental evidence. I understand it is difficult to conduct experiments, but if that is the case, stop being so certain and adamant in your position without evidence

Nit: observed values converge on expected values over many trials. If you asked me which strategy I'd like to play 100,000 times, I'd take the 50% chance. Just once, I'll take the $49.

"A large amount of individual debt is wholly optional."

And a lot isn't.

Apart from my mortgage, ALL of my debt is accrued when I'm too sick to work, and the medical bills pile up.

Some huge fraction (1/3?, 1/2?) of bankruptcies are caused by medical related expenses.

It may have been that debt that ended up on the books, but what do your monthly expenditures look like in terms of vehicles, eating out, entertainment, etc? If you have cash on hand to pay those expenses you have cash to pay your debit. This may not be your personal case but I have seen this a lot.

Sorry for delayed response. I had to reread your comment to understand.

My burn rate is very low.

When I'm not working, it's because I'm sick, meaning no income. I'm too stubborn or principled or stupid to take unemployment (unless I'm actually looking for work). And I'm not sick enough for disability.

During those times I chew down my savings. This last round, I dipped into my 401k. (I had 6 months savings, expected to be out of work for 2 months, stretched to 10, had some major expenses come up.)

Then I get somewhat healthy, pay off my new medical bills, save what I can. Until next time.

Rinse, lather, repeat.

See Fisher's theory of debt deflation: https://en.m.wikipedia.org/wiki/Debt_deflation

The median household probably has to pay taxes too.

Reply on news.ycombinator.com