A model for economic mobility

Yeah, I did the immortal thing myself before I got married, but I paid the price when my patelar tendon severed. Surgery, hospital costs, rehabilitation == ouch for the uninsured. Still, I wouldn’t wish my luck on Sam.

Speaking of marriage, though, it was a significant expense to me. We did things relatively cheaply, but wedding and honeymoon still set me back about $7K (most of that was honeymoon, now that I think of it. Say $2500 for the wedding & reception, thanks to some sweet deals with a UU retreat center.) Are we assuming that Sam and Sally are already married at 18?

As for Sam’s income, I’ll trust you to know what is reasonable for your geographic location. I’ve lived in towns where $50K put one well into upper middle class, but where I live now anything les than 6 figures gets you sneered at in the valet-serviced malls. (I’ve gotten used to it. Sometimes I scratch myself just to see the looks of horror. ;)) On the national level, though, $50K of taxable income placed one close to the top quintile according to these 1997 numbers (the FEI is Treeasury Department’s Family Economic Income – not relevant ot our present discussion but interesting in and of itself.) That’s taxable income, not gross, but after exemptions I would imagine that Sam’s final salary still places him around the 60% mark.

OK, I’ll bite. First, I’m betting they take out a 30 year mortgage on their first house. There’s no way they will think about buying the second house until that’s paid off.

You’re being generous that they will be married and liquid at age 18. That’s assuming substantial parental help I believe. Don’t know if that’s typical. I think not. IIRC, lower income/working class people tend to marry earlier and likely to come from the working class. I won’t pick this nit, but find it unrealistic. Will come back to this point.

Raised earlier on how to model windfalls/disasters. I believe the latter is much more likely, and that the two factors will not cancel each other out. Your basic windfall is parents passing on and leaving an estate. Again, I’m assuming that the couple are from a basic working class background. Do you agree? The bulk of the estate is at best one working class house to be split among all siblings. This also assumes that the parents did not get divorced and divide property, and if they did, since divorce rates are pretty high, the estate is going to be paltry. Against that is the risk of being laid off, a market crash, emergencies, a serious medical event that outpaces insurance and in the worst case leaves the breadwinner out of the work force for an extended period of time, natural disasters, etc… IMHO, the risk of a bad event is much higher than that of a good event. How do you model such a probability?

I will be constructive here. Get your basic model down. I won’t pick nits, although I wish posters would say if the returns are inflation adjusted. At the end, we also have to present value your analysis. Now, here is the kicker, once the basic model is accepted, then run a sensitivity analysis. What happens to that retirement account when there is a major medical emergency? Or a car dies and must be replaced? The car dies in year one versus year 30? Or those happy newlyweds start out broke or in the hole? A divorce happens halfway through. Hell, I’m not a pessimist, you could have a windfall inheritance or win the powerball scenario as well. You could really get tricky then and assign probabilities to various scenarios and get a mean return.

IMHO, what you will find is that it isn’t as easy as you think it is to parley a couple hundred dollars liquidity a month into a house, college for the kids and retirement. I’ve run the time value money calculations as well, and theoretically if you can in fact put a little away every month it does become substantial. But, okay, let’s see how this game plays out and then compare it to real life.

Let me second China Guy’s comments. In particular the expected value = 0 for windfall/neg windfalls strikes me as exceedingly unrealistic.

I, too, have to argue for the ‘disaster’ outweighs ‘happy accident’ part of the model.

S&S are, in the beginning, like a poker player coming to a game with too little cash. One early bad break can send them into bankruptcy. Contrariwise, getting a windfall won’t have any real long-term consequences. No one sends lawyers after you (carnivorous ones) for having too much money…only for too little.

So you’re going to have to assign an probability to a economic disaster or windfall occuring each year of the model and see what happens. Do it that way and you also account for a ‘run of bad luck’ where S&S are just cursed. Disasters don’t promise to just occur once.

I’d say you’d have to:

A) Determine the probability of windfall and disaster
B) Have a formula for determining the extent (ie Windfall: win lottery for $1500 or $100,000,000 or Disaster: Transmission goes spend $2K or run over a lawyers child pay $1,000,000 and lose everything)
C) Run that algorithm for each year of the model

Man, I love this stuff. Let me know if I can help. This one should be interesting.

What I intend to do is run the analysis clean the first time through: no windfalls, and no disasters.
I want a baseline run.

Once I’ve done that, I can input it into some software and very easily run some what if scenarios for various disasters and windfalls at various times.

If his income at peak earning time, after thirty years of employment only gets him to the 60th percentile, I’d consider that mediocre enough. In other words it sounds good to me if it sounds good to you.

***We’re doing things in today’s dollars and adjusting returns for historic inflation of 3%. For example, the S&P earns 11% in hard dollars, adjusted for inflation it’s 8.

As for the starting off point, I think it’s modest and realistic.

Here’s their story:

Sam and Sally have been sweethearts forever. Outside of shop class in Sophomore year Sam proposed to Sally and gave her a ring made of sheet metal.

"You’re my main girl Sally. I may not be much, but stick with me. We’ll work hard and save money. Next year I’ll buy that Iroc Z I’ve always been talking about with money I’ve made pumping gas. Then we’ll save up some cash, get married after graudation, have the reception at the fire hall, and have enough left over to start our life.

I may not be smart, Sally. I may not be talented, but I’m a hard worker with a dream of a family, an IROC and a mullett haircut.
Stick with me and we’ll raise the perfect family and retire rich.

Please say “yes” Sally. Don’t break my heart, my achey breakey heart."

Sally said “Yes” and somehow it stuck. Sam worked at the gas station, while Sally sold sea shells down by the sea shore, and they had the wedding, the reception, the whole nine yards.

Since they can’t count on any substantial help from the broken homes they came from, I don’t feel bad about starting them off with $3,000 in cash, an IROC and $1,000 worth of personal property.

I’d had more than that that I’d eaned at 18.

Have things changed? When I did this for myself on my 401k I assumed an average of 10% per year nominal for the S&P, and 7% adjusted for inflation, using the Ibbotson numbers that start from 1925. Any reason for your 11/8 assumption? That 1% can make a huge difference over time.

A quibble about your original assumptons:

**10. He would like to retire at age 60, but can work until age 65. **

Current US law is, I think, that he can continue to work until physically unable to do so. So he could work significantly beyond 65, if he so chose.

Well, if I were going to quibble with return rates it would be for “conservative” investments after 50 which still yield 5% above inflation. But I’m happy to run the more generous nubers once and then play with the conditions and see how the result changes.

Scylla, I’m not going to make an issue of your beginning conditions or Sam’s upper-bound salary, but I do think we should remove the label “underachieving” from our description of the poor chap. If at 18 he owns his car outright, has 3K in the bank plus another 1K in real assets (that hold value?), and has a full-time job that pays substantially above minumum wage–well, that doesn’t sound particularly underachieving to me.

Dammit, I don’t have time for this thread. I’ve got exams in the next few weeks!

As such, I’ll restrict myself to one point: 8% real yield on equities in the next thirty year time frame is hopelessly optimistic. The past is not necessarily a guide to the future. In high inflationary times, interest rates can easily massively outpace inflation (since they are the primary control on inflation) and refuges from inflation, such as equities, get bid up. This has been the case for most of the last century, due to a few large oil shocks and the recovery from the second world war.

Current financial models (in the UK at least) are using a 30 year or so real yield on equities of about 3-4% and real yield on FI bonds of about 1-2% (IL bonds are generally less than 1%). Add on 0.5% for AAA-rated coroporate bonds.

I think you should be using about 4% real yield on equities maximum. To do otherwise would be out of line with other financial models in the market.

(Anyway, think about it: growth in equities over a suitable time frame (say 50 years) should be the same as the growth in GDP over the same period, so long as the division between land, labour and capital remains static. Do you really think that you are going to get 8% real GDP growth over the next 50 years?)



While it’s true that past performance is no guarranty of future returns, using a long term track record as a predictor is IMO a superior method to trying to guess what the markets will do for the next thirty years.

My 11% adjusted for inflation to 8% is quite realistic, as performance figures for the SP500 are available going back to the early 1800s, and is based on last year’s Ibbotson’s data. If anything, I think that we’re overestimating inflation do to some recent abberations, but since the mean is 3%, that’s what I’m using.

There is a very strong case that real economic growth has been hampered in the past 100 years, and that we really do have great expectations looking forward. Technology has greatly speeded up commerce, we have the emergence of a world economy, and the potential contribution of the third world to look forward to. We have the resergence of Asia, and Russia entering the free markets.

If we figure things historically, a large perdcentage of the return of the SP500 is dividends, about 3.5% if I recall correctly. That means are number for stock market growth is 7.5% and down to 4.5% in real terms. It’s quite conservative, especially when you consider tha their are funds out there like the Income Fund of America that have total returns of 13.5% over the last 70 years with diversified portfolios of stocks that historically grow their dividends.

If you wish to make the argument that a broad selection of large cap stocks historically and in the future have and can only be expected to return 4% in real terms, please do so. I’m eager to hear it.

Until you make the case, I feel quite justified in using the historic numbers. They may be only half sighted, but that’s better than the a blind assumption (and actuarial projections have not been especially good indidcators of market performance in the past, no offense.)


Perhaps we’re quibbling too much at this point. It’s been my experience that quite a few teenagers work during high school and frequently purchase their own cars and acquire assets. This is more likely to occur in rural areas with kids that aren’t going to be going to College but are proceeding directly into the workforce.

His top income at age 50 is going to place him in the 33 percentile for income, and I just pulled that from 5 year old data, so it’s probably not as good now.

I chose it that way so that Sam would be spending roughly half his working life below the mean, and half above, and capped his income at 50k so we wouldn’t inadvertently push his income too high.

In reality he’s quite the underachiever, as mean household income numbers are for all households in the US, not just those with members gainfully employed. After working 30 years, he should be doing better.

So, I feel pretty confident that Sam is on the low side of mediocre.

To be fair, I really am giving this couple one major untypical advantage. I’m starting them off a little young.

Once we got it done the first time though, we can easily play with it, and start them off at age 23 for example, or make it a shotgun wedding and give Sally a kid at 18.

I think the assumptions are fair enough that we can fiddle with thme after the first run.

Let’s start budgeting the first few years, 18-25. Let’s hash out their expenses. I’ll see what people post, and try to put it together this afternoon or evening.

Scylla, if you want to use 8% real yield for the purposes of your model, that’s fair enough. It’s your party after all.

If you tried to use it in an actual model for the purposes of valuing a pension scheme, life office or long term insurance liabilities however, you’d be laughed out of the board room. If you tried to justify it by pointing to data from the 1800s like it has any relevance whatsoever… well. Put it this way - they’d probably stop laughing and start phoning the men in white coats.

What does stock price and income depend on? Start identifying that and you start to see why the results from the 1800s have little if any impact on financial futures in the 21st century.

Geez - even when we were being incredibly and naively optimistic in our long term assumptions 5 years ago we were still only using 4.5% real yield on equities.

Oh - and you have allowed for management charges I hope? That’ll knock 0.5% - 1% off the yield per annum. And I doubt your investors get their income tax free, so don’t forget to knock the basic rate of tax off too. And the fact they won’t be fully diversified increases their exposure to bad debt - you should probably knock another 0.25% off for that too.




I’m looking at the Ibbotson numbers through year-end 2000 right now (they haven’t published the 2001, or if they have they haven’t sent it to me yet, and I purchase their research.)

Going back to 1925, the number for large cap stocks is 11.0%

I hate to rain on your parade, but the performance figures I quoted were net of sales charges and expenses. Most SP500 funds out there use derivatives to offset their expenses and more closely mirror the SP500, and do so to a pretty decent degree. Even if they didn’t I wouldn’t be bothering to factor it in because it would be a colossal pain in the ass to do so.

As for your actuarial assumptions about growth; no offense but the actuarial assumptions used in pension management are in fact an effective contrary indicator for the markets’ performance. i am certainly not the first to think so. It’s a large issue as far as Mr. Warren Buffet is considered as well as Fed Chariman Greenspan. It’s an issue as well in our current accounting debacle.

Basically, the problem is that if equities do well for a year or two actuarial assumptions for equities go way up. If they go poorly, they go way down. Works the same for bonds and interest rate assumptions.

So, again but actuarial assumptions about future market performance are perhaps the worst single tool you could ever use to predict returns. In fact, as you should well know, they should not be used in this fashion.

Finally, your theory about inflation boosting stock prices is absurd. Chart inflation vs. equities for the last hundred years and there is no correlation. If anything there’s a negative correlation. I don’t pretend to have a crystal ball, but I sure as shit know you don’t either.

So, I’ll just assume that the conditions that have prevailed for the last 150 years will continue to prevail absent a compelling reason to suggest that they shouldn’t.

Nothing else would be reasonable.

This is a incorrect. Read the fund prospectuses. Some but not all funds are allowed to borrow, leverage, loan stocks and invest in derivatives in order to increase potential returns. Returns off of things such as stock loans and derivatives are part of the total fund return (or loss). Funds do not invest in derivatives and then take those returns (or losses) to pay fees and expenses.

Funds invest in derivative products to quickly mirror an index or stock, gain gearing, and reduce risk or increase returns. For example, funds with large stock holdings may sell upside calls if they thing the market is flat or declinng (buy write/covered call), and they will outperform if the market falls, and conversely underperform if the market rises.

Index derivatives may be used to mirror the S&P 500, but do not more closely mirror the S&P500 than being long the exact basket of underlying stocks. Correct me if I’m mistaken that the S&P futures are cash and not stock settled. Most derivatives on the S&P 500 are based on either the index or comprised of tracking stocks. Index derivatives may outperform the undelying index owing to less transaction costs and messing with dividends. However, cash settled futures on an index are only equivalent and not the same as the underlying.

Back to the OP, again I would suggest accepting Scylla’s basic model. Once that is run, then you can do the sensitivity analysis with a 4% inflation adjusted return, braces for the kiddies, car accident at age 25, etc.

I have to agree that the long term historic real stock market returns are a likely indicator of future long term real returns.

China Guy:

I was quoting returns on a specific fund I named and those returns were net of expenses. As for SP500 funds; in reponse to the performance lag in many index funds, we’ve seen a new crop of “mirror” funds whose goal is not to simply purchase the stocks in the underlying index but rather to come as closely as possible to mirroring the index as a whole net of fees.

You buy one of these no-load funds and you’re supposed to get as near as possible to exactly the return of the SP500. They’ve been around 5 years or so, and tend to a pretty good job, hitting the number within 10 basis points or so, plus or minus, year after year.

Index funds come in three flavors as far as this is concerned:

  1. Own the underlying stocks.
  2. Own derivatives
  3. Hybrids that own both

So, basically what I’m assuming is that they buy a fund which mirrors the SP500 so close that it doesn’t really matter. One can buy such funds. They exist.

The vagaries of what fund they buy are kind of beside the point since we’re assigning them a rate of return based on an assumed rate of return.

How about we get back to the point, get this baby off the ground, and then we can fiddle with the assumptions?

Here’s what I came up with for a budget for the first period for Sam and Sally.

If I put an “*” next to the item, that means it’s an actual cost that I figured out, otherwise it’s just an estimate.

I’ll leave the list here to be haggled over for a day or so, before we run the first few years.

Mobile Home rent (doublewide, fair condition, rural location) - $225/month*

Electricity -$60/month*

Gas heat, prorated -$47/month*

Auto insurance (IROC, Pa, 18 year old) - $675/year

Clothing - $500/year

Phone - $50/month

Gas for the IROC, figuring a tank a week - $80/month

Sally’s medical visits - $160/year

Sam’s medical visits - $90/year (see earlier post,)

Sundries - $200/mos.

Car Maintenance - $50.00/ mos

Unnamed home costs - $35/mos

home purchases - $50/mos (this would be things like toasters, coffe makers, furniture and other necessary appliances.) Rather than bill them all at once the first year, I figure I’d spread it out some in recognition of constant replacement of worn out things)

Home set up costs - $1,000 one time only, in the first year. I’ll assume they got married, had some stuff, got some other stuff for the wedding, some junk from parents and yard sales, and such, but when all was said and done, they had to spend another grand to get what they needed to set up their home.
Any thoughts, comments, additions, subtractions?


Renter’s Insurance?

The auto insurance seems low for an 18 year old with a sports car, even in a rural area. I tried to look up an online rate quote, but they are balky about giving out quotes for rural areas, it seems. Know any insurance agents? And I bet Sam Sample has had at least one speeding ticket with his IROC.


When I moved down from NY to here, my straight liability 300/100 went from over $1200 to $400, and as I’ve gotten older, it’s gone down since.

Then again I was 25 with an old jeep, not 18 with an IROC. That’s how I made the guess.

I’ll call my agent tomorrow, now that you’ve piqued myh curiosity.

I got the figures for the double wide by tlking to a girl who works in my office who lived in one up to last year.


Well, I just got off the phone with kay my insurance agent and got Sam an actual quote (which was hard to do without a SS#, since running a credit check is part of the application process and partially dedtermines premium.)

Anyway, the number is $1902 a year :eek: :eek: :eek: :eek:

Holy Gaphoney!

That’s more than I pay for three cars with collision.

She said the fact that it’s an IROC and the fact he had a speeding ticket are only about $200-$250 worth of that number.

It’s being 18 that’s the sin.