## The challenge of fitting “good advice” into a coherent course on statistics

Let’s also not forget the benefit of the occasional dumb but fun example. For example, I came across the following passage in a New York Times article: “By the early 2000s, Whitestone was again filling up with young families eager to make homes for themselves on its quiet, leafy streets. But prices had soared. In October 2005, the Sheas sold the house, for which they had paid \$28,000 nearly 40 years ago, for more than \$600,000.” They forgot to divide by the Consumer Price Index! Silly but, yes, these things happen, and it’s good to remind social science students that if they know about these simple operations, they’re already ahead of the game. The next step is to discuss more difficult problems such as adjusting the CPI for quality improvements. (For example, if the average home today is larger than the average home 40 years ago, should the CPI adjustment be per home or per square foot?) I also like to mention points such as, “The difference between ‘significant’ and ‘nonsignificant’ is not itself statistically significant.” But I haven’t surmounted the challenge of how to fit this sort of “good advice” into a coherent course so that students have a sense of how to apply these ideas in new problems.

Maybe we’re not all the way there yet, but we’ve been working on it. In Regression and Other Stories we attempt to integrate good statistical advice into a coherent course on applied regression and causal inference.

1. John Hall says:

I’m not saying you shouldn’t adjust for inflation here, but even that makes doesn’t provide that much information. What really matters is how investing in the home would compare to investing in other assets. Something like internal rate of return is probably better for that, but you would still need to compare to what if they rented and instead put the same amount of money they spent on mortgage/taxes in stocks (or a diversified portfolio). That’s the economic decision.

• Carlos Ungil says:

What matters depends on the context. How investing in that home would have compared to investing in other assets is quite irrelevant. I cannot read the article but https://www.nytimes.com/2007/09/02/nyregion/thecity/02aspi.html doesn’t look like investment advice. The article gives factual numbers: many adjustments could be made and the CPI may not be the best one. Real per-capita GDP more than doubled in that period and maybe a more relevant denominator would be something else (like the median household income in Queens).

• AllanC says:

+1

I think John is thinking about how well they did in their choice of investment vehicle (where a comparison to other asset classes and risk-adjusted returns are important).

But the article seems to be travelling down a different path. Based on the quote the author appears to be using the fact that the nominal value of the home increased by more than an order of magnitude over a 40 year period to suggest that prices have increased outstandishly. In which case, it isn’t a comparison to other assets that is of concern so much as it is relative affordability. How one defines relative affordability is probably up for some debate but certainly income available for housing / housing prices may be a decent indicator (or the inverse).

• Martha (Smith) says:

I’m not a mom, but I’m plenty old enough to be a great-grand-mom, so I’ll try to give some input that I hope will be speaking for the moms.

A house isn’t just about financial considerations. What you pay for a house is in part like what you pay for a vacation: Pleasure, enjoyment. Just his morning, I spent several minutes looking out the windows at the flowers (on trees, shrubs, from bulbs, etc., etc.) in my yard. It would be very unlikely that I could have that experience (just by looking out my window) if I lived in a rental house. Another part of the enjoyment is making the house (as well as the yard) what you’d like it to be, with additions or remodeling or whatever. And for people who are grandparents or great-grandparents, there is the enjoyment of having the grandkids or greatgrandkids come over, and looking over old photo albums or whatever, or telling them when you planted the trees, shrubs, or flowers that they and you like so much, etc., etc.

• jim says:

“What you pay for a house is in part like what you pay for a vacation: Pleasure, enjoyment.”

Yes, I agree.

2. jim says:

I’m not sure if you need to compare to the CPI in this context.

My mom gushes all the time about how much the house she bought in 1974 is worth now. She doesn’t care about it’s value relative to the CPI.
She doesn’t care that the return on the investment is probably negative after you take into account all the taxes and refrigerators and roofs and fences. She doesn’t care that, as long as she’s living in it, it might as well be worth nothing because she can’t eat it or fly it to Hawaii for a vacation.

All she cares about is that it’s worth more than ten times what she paid for it! She made a great investment! That’s thrilling to most people.

I think it’s important. Ignoring all the stuff you mention in your post leads to people pouring all their assets into an illiquid, undiversified asset (their house) because they think it’s a great investment. Some rationality about real estate would tamp down speculation, cool off overheated housing markets and may even be good for the environment if it reduces the number of McMansions.

That said, I don’t think your mother’s house “might as well be worth nothing”. She’s living rent-free. Sure, she has to pay all the costs, but if she were renting a comparable place, the landlord would pass all those on to her, with a markup.

• Kyle C says:

As someone with a similar mother, I know just what Jim means. First, Adede, if Mom rented a place, it would almost certainly be newer and smaller than the house and thus require less maintenance. That’s a saving right there. And if someone who is, say, 85, spends some of their house equity on rent and some to fund nice vacations, they can have a better life at virtually no financial risk, given their life expectancy. Older people can get (1) inflation blind* and (2) very attached to feeling rich because of what they think** their assets are “worth,” which as Jim notes is all on paper and might as well be nothing for all the good the asset “value” does them.

* my mother still in good faith gives her grand and great-grandkids \$1 for each year of their birthdays, even though this is ~ .1 of the value when she was a girl and .2 of the value when her kids were young,

** not just the house but the Hummel figurines and that dumb painting they found at a flea market that “could be worth something.”

• Ben says:

I’m still kinda with jim’s mom on this one.

> return on the investment is probably negative after you take into account all the taxes and refrigerators and roofs and fences.

Adede’s point seems like a good one:

> but if she were renting a comparable place, the landlord would pass all those on to her, with a markup.

I guess it’s possible landlords have been biting the bullet all these years and not forwarding the bill, but I’m skeptical. They’d go bankrupt! To the point about eating, your mom also can’t eat the money she paid in rent or fly to Hawaii on it.

So there’s property tax, there’s inflation, presumably the house was financed so she’d have paid interest on that, and there’s some amount of dollarydoos per year spent on repairs to major appliances and parts of the house. Those would be the things going into the cost of the house. Is there anything else?

On the other hand there’d be the rent, which jim’s mom would just lose. Are we assuming she throws any remaining cash into the stock market or retirement or something?

And all of this is over time so there’s the whole inflation thing.

• Ben says:

Actually I forgot to list the price of the actual house lol.

Also the liquidity thing is weird — isn’t that what mortgages are? You can convert the house back to cash instantly at some cost, in which case you can eat the house and fly it.

Anyway as a renter I am happy to be wrong and find out this all has been such a great investment.

Houses are illiquid. Getting cash from a bank account takes 5 minutes at the ATM. Converting stocks to cash takes one business day. Selling a house can take anywhere from months to years, depending on the local market, economic conditions and how much you hope to get.

Taking out a mortgage isn’t converting the house to cash. It’s taking out a loan, which happens to use the house as collateral. I suppose you could get a comparable loan with any other asset appraised at a similar value. I heard of a taxi company owner who got millions in loans using his taxi medallions as collateral. When Uber caused the value of the medallions to plummet, he was in a much worse position than if he had just sold them.

If you sell an asset, you may get less money than you’d like if you price it low to expedite the transaction, but you can never end up with less than \$0. If you take out a loan on an asset and spend the money, it’s possible to end up with a negative net worth.

• No, you’re not with Jim’s mom, you’re talking about things like inflation and the comparison between buy and rent… Jim’s mom is looking at the nominal purchase price and the nominal sales price and ignoring all that other stuff.

The right way to evaluate a house is to compare the lifetime values

sales price – purchase price – capital gains tax – all the maintenance costs – all the maintenance time – all the mortgage payments + the equivalent rent, all discounted according to the time-varying discount rate depending on when the cash flow occurred.

It’s actually a fairly complicated analysis, which is why most people, who have sort of barely passed HS Algebra or something haven’t a clue how to do it. I think about half of typical retail money management advisors couldn’t really set it up right either.

• In terms of financial considerations the place where Martha’s concerns (from https://statmodeling.stat.columbia.edu/2020/09/18/the-challenge-of-fitting-good-advice-into-a-coherent-course-on-statistics/#comment-1488861 ) enter in is the “equivalent rent”.

In other words the value you get out of living in the house reduced to monetary terms is the amount of money you’d be willing to pay to live in your house and have all the stuff/garden/configuration be the same, if someone else owned the house and you had to pay rent to them.

What the “market rent” is, isn’t as important as your own personal willingness to pay. For example if you really like having the walls painted eggplant colored, but the typical person doesn’t, then you’re willing to pay more than the market rent to get that eggplant color.

• John Williams says:

This discussion seems more about Homo economicus than about people. If economics is going to be a science, it needs to be about what people do, not what economists think they should do.

• I disagree. Both concerns are relevant. I mean imagine if the study of human health was all about diseases that exist, but not at all about how to prevent or cure them?

Suppose you randomize two large groups of people. One you simply do annual surveys on their finances and happiness, the other you give them a course in how to make decisions using these kinds of techniques… 20 years later you find the people who got that course are wealthier, healthier, happier, suffered fewer bankruptcies, had more career satisfaction, etc.

Would that be irrelevant?

• Martha (Smith) says:

John Williams said,
“This discussion seems more about Homo economicus than about people. If economics is going to be a science, it needs to be about what people do, not what economists think they should do.”

+1

• Another way to put this is that Economics *isn’t* a science. Economics, like medicine, is a branch of engineering. When done well, it’s about finding ways to improve people’s lives through better decision making about use of scarce resources.

This is actually why I am so disappointed in economics as practiced today, because I think it often repudiates this core obligation it has.

• Scuba Diver says:

>Economics, like medicine, is a branch of engineering.

I think this is a not so useful way to put it. Engineering is most of all applied science. Engineering has theoretical backing in multiple “pure” sciences as does any applied science. Economics however… What’s the home base of economics? Social sciences? Doubtable, there seems to be a firmly rooted set of prior believes in economics, that is not generally accepted in social sciences.

• jim says:

“I guess it’s possible landlords have been biting the bullet all these years”

Landlords don’t need to “bite the bullet” for apartments to cost less than homes. Aside from having a (dramatically) lower square footage than a home, and aside from mostly not having any yard, an apartment building has far lower costs per unit area than a house, and those costs are shared among all the renters to boot. For example, an intelligent landlord might replace all appliances every 10 years, get a bulk discount on the new items, reduced maintenance, then sell the older appliances on a secondary market.

Property tax is a big savings in an apartment. My mom’s property taxes have more than doubled since 2013. Yet a home is a depreciating asset. So most of that increased value is in the value of the land. In an apartment you obviously consume less far land: your living area is smaller, but also you don’t have a yard. But even better: you also subdivide your 1200 ft2 with the people living above and below you.

Not to mention the fact that the cash that is now in the house could earn a substantial investment return. Even a very conservative investor could squeeze 3% out of the cash (\$500K home would generate \$15K a year). But a smarter investor might put \$100K in cash against emergencies, then put the other \$400K into higher return assets that could pull a cool 10%.

I mean, that’s just evaluating a home as an investment. Obviously people have other reasons for owning a home. But also the cost dynamic changes as ppl age, so if that’s a concern it has to be reevaluated as time goes on.

You keep saying that renting a small apartment costs less than owning a house. Which is true, but it misses the point. By comparing radically different types of property, you’re introducing a confounding variable (just like comparing house prices over time has the confounding variable of inflation). Yes, renting a 1200 sqft apartment is cheaper than owning a 3000 sqft house. But owning a 3000 sqft house is cheaper than renting a 3000 sqft house. And owning a 1200 sqft apartment is cheaper than renting a 1200 sqft apartment.

• Yes, but a big part of the crazy high housing costs is that single old ladies are living in houses they bought 50 years earlier when they had a husband and two kids, but because of capital gains taxation and property tax breaks (particularly in CA with Proposition 13) they stay in their homes because it’s actually more expensive to sell the house and move to an apartment (after paying “capital gains” on the **nominal** value increase of the house). So one person is living in a 3 bedroom 1800 sqft house, and running it into the ground with insufficient upkeep instead of moving to a 900sqft condo, and a family of 4 are squeezing into 900sqft condos.

part of the point is that homeowners, because they’re making decisions that don’t really look at the proper questions, plus bad structural incentives, are choosing to do things that are highly inefficient and harming young families in the process while reaping govt rents.

Of course, I forgot about perverse tax incentives. Anyone who can keep track of 50 sets of arcane regulations should be giving advice, not taking it from me…

• jim says:

“owning a 3000 sqft house is cheaper than renting a 3000 sqft house. “
“owning a 1200 sqft apartment is cheaper than renting a 1200 sqft apartment.”

I don’t think you can say either outright. Like any problem you need to make your assumptions explicit: how long will you rent/own that particular property? Are property values rising or falling? What are the interest rates on home loans vs. return on other investments? IE: what are the alternative uses of your cash?

From a purely investment standpoint, there are many factors that would go into that decision. It’s not as cut and dried as you make it.

• jim says:

Daniel said:

“it’s actually more expensive to sell the house and move to an apartment (after paying “capital gains” on the **nominal** value increase of the house”

Wow is that really true? CA has so many bad incentives! OMG. It’s economy is living on borrowed time. Sadly Seattle and WA are copying CA.

• Jim, it’s true, but it’s not just a CA thing, a bunch of this disincentive is Federal. Federal tax will hammer you with capital gains, unless you die, in which case you can have your heirs inherit the full value of the property (up to something like 5M).

Yes, you can roll your capital gains over into a different house, but it means you have to buy a house with the same value or more if I remember correctly. So it lets you move from place to place, but not to downsize.

Where it gets worse for CA is with Prop 13. Most people who bought a house 50 years ago are paying 1/10 the property tax of people who move in next to them. There are some circumstances where you can roll this over to a new location, but it’s got complicated county-to-county reciprocity agreement issues, and it just basically sucks.

The thing is Prop 13 isn’t some new “liberal plot” or anything, Prop 13 is a 1970’s damn fool rent-seeking anti-tax GOPer type plot. Most of the modern “liberals” I know are against it because it ruins public schooling in CA. In fact there’s something on the ballot trying to repeal it for the purposes of commercial real estate, which hopefully will pass.

So it’s complicated.

• jim says:

“Federal tax will hammer you with capital gains, unless you die…”

I just looked up Federal. It says you can exclude up to \$250K (single person) and \$500K (joint filer)! So as soon as my dad died my mom’s home was subject to \$50K cap gains tax? That’s an outrage! I thought you could exclude the entire value of your residence. Must be brutal for people in CA, where property values have gone up SOOOOO much.

Of course if you count in the mortgage tax deduction you probably still beat the government, but it’s bullshit that the tax is hidden, and it’s also bullshit that the more frequently you move and sell (ie., you have less gain to tax), the better tax deal you get).

• Jim’s real point is that the vast majority of people are TERRIBLE at evaluating financial questions. Truly TERRIBLE.

I mean, just the idea that you could take a set of future payments, and a return on a bond investment, and turn that into a net present value (ie. the amount you could borrow if you were willing to make those future payments)… That’s like advanced math that maybe 97% of the US can’t do.

• jim says:

“the vast majority of people are TERRIBLE at evaluating financial questions.”

Ha, so funny, a few years back my mom bought a car. She got a zero percent loan but I have to keep stopping her from paying it off!!! Too funny.

• Martha (Smith) says:

Who gives zero percent loans?!?!

• jim says:

Martha (Smith) says: “Who gives zero percent loans?!?!”

Lots of car dealers! For sure! The game is to hook you on a low- or zero-interest loan then upsell you on extended warranties, maintenance contracts etc.

• Martha (Smith) says:

Ya learn something new every day. If I remember correctly, I’ve owned three cars in my life, and all of them financed with loans through a credit union. Never considered financing one through a dealer.

• jim says:

Well it worked out great for my mom. Instead of spending \$30K cash (her plan), she took the no interest loan. The cash went into a fixed income investment that yielded ~6% dividend plus a little cap gain and she scored about \$6,000 on it over the life of the loan to give to grandkids.

• Jim. This is the kind of thing I try to explain to family and friends all the time. They often just don’t get it. it’s too unintuitive to them, and they don’t have the educational background to understand the calculations. It’s particularly true when the situation becomes more complicated than just a car loan (like, when should you retire, and should you cash out your retirement or leave it in the pension plan? should you pay off your mortgage or not? should you sell your house if you want to move or not? how do these decisions depend on your age? etc)

• One of my favorite calculations like this was that I showed my wife it made more sense to scrap our working washer and dryer that came with the house (they were circa 1985) and get new Kenmore front loaders switching the dryer from electric to gas… the reduced cost of energy would pay for the cost of the new appliances within a couple years (I don’t remember the exact numbers but a major part of it was not using electricity for the dryer) and the newer machines were much more convenient and functional.

• jim says:

“they don’t have the educational background to understand the calculations”

I think a lot of people, even very highly educated people, just don’t believe in things that can only be shown with calculations.

One of my favorite financial tropes is the “age-risk balance” philosophy – the widespread advice given by financial professionals that you should shift your investments into low-return fixed income as you age. It makes no sense for anyone. The returns of the broader market are so much higher that they offset any value that might be lost in a downturn in only a few years. Yes, the LTAvg return of hte market is 10%, but that includes depressions :) Boom years are typically 25-30%.

• Jim. I’m literally having that conversation with my wife this week. yes you might want to shift sufficient quantities into stable investments that your cash flow doesn’t become pinched so you don’t lose your house or something like that… But as long as you can make 5 years of house payments (or you own your home) using your social security and some fixed income returns, having the rest in stocks makes good sense even for 70 or 80 or 90 year olds. In fact, as you age, it makes even more sense, as you should be investing for your children and grandchildren, provided you aren’t on the brink of financial collapse…

Of course the fraction of the US that *is* on the brink of financial collapse by age 70 or 80 is high.

We should join a club, get some t-shirts.

• jim says:

“But as long as you can make 5 years of house payments (or you own your home)…”

Well I would say two years, but yep, that’s the idea. Also I would stick with cash for the cushion money.

You know when I hear Krugman say “well the stock market isn’t the economy, millions of Americans don’t own stocks!” I cringe. It’s not a good thing that they don’t own stocks. It’s a bad thing. It means either a) they have no excess income to invest, or b) they’re financial ignorant and putting their savings in banks. Lots of immigrants are in the (b) camp, and that’s really sad because they work so hard for so little and save so much, but they earn nothing on their savings so in the end they have to go back to their home countries to retire, unless their kids do well.

• Here’s one where the typical investment advice gets it far wrong:

It’s “tax Free” to trade in your IRA/Roth IRA/SEP Ira/401k/403b/etc whereas if you want to rebalance your regular investment accounts you have to immediately pay capital gains taxes on proceeds.

What this means is that you should buy relatively more stable long term investments in your brokerage accounts, and use your IRA/401k/etc as the place where you trade things in order to rebalance. So for example suppose your brokerage account has an investment in a small-cap index and small caps do very well… it becomes a bigger portion of your portfolio, you want to reduce your *total* risk of small caps, so you sell your all-cap fund in your IRA and buy a large cap, thereby rebalancing your portfolio, tax free.

But if you go to an investment advisor they’ll be telling you to hold a bunch of bonds in that retirement fund because it’s all about capital preservation for when you’ll be living on a fixed income after you get your gold watch for 40 years on the assembly line at Ford Motors or whatever. It’s terrible heuristic advice from an age gone by.

• Carlos Ungil says:

> The returns of the broader market are so much higher that they offset any value that might be lost in a downturn in only a few years.

Not 100% sure. Depends on how many years are a few, how (un)lucky you may be, and how much the future looks like the past.

If the starting date had been 1999, stocks would have underperformed government bonds until the second half of the 2010’s decade: https://www.etftrends.com/wp-content/uploads/2019/05/Stocks-versus-bonds.png

That chart is for 10yr bonds, long-term bonds are still doing better than equity twenty years later. (Disclaimer: Past performance etc.)

• jim says:

“Depends on how many years are a few, how (un)lucky you may be, and how much the future looks like the past.”

What you say is true, but it presumes that at some starting point (1/1/2000?) you chose all stocks or all bonds with all of your money. In reality people invest gradually over a life time, so that model isn’t relevant. Under a gradual investing scenario, it’s just a matter of the long term averages, which strongly favor stocks.

• Carlos Ungil says:

Sure, but the long term may be too far away if you know (or suspect) you will need the money in a few years. These considerations may be irrelevant in the gradual investing scenario, but not so much in the gradual divesting scenario.

• jim says:

“Sure, but the long term may be too far away if you know (or suspect) you will need the money in a few years. “

You shouldn’t have to face the divestment issue in the way you describe it. Keep at least 10% of savings in cash all the time. When you’re younger, you just don’t invest everything; when you’re older, you feed the cash pile when the market provides an opportunity.

• Jim,

Median household net worth for black families is somewhere in the \$5k to 15k range. So there it’s clear the median family is impoverished. The kinds of numbers we’re throwing around here don’t apply to their situations.

For white families it’s like \$150k. But even \$150k if your mortgage payment is \$2k is only 6.25 years of payments. So keeping say 2 years in liquid low risk investments (say cash) is \$50k = 1/3 of your assets. If your house is half of your net worth. then the \$50k is 50/(150/2) = 67% of your liquid assets.

I think the fact is far too many americans don’t save enough, don’t invest enough, don’t make very good financial decisions, and wind up far poorer than they needed to be. Possibly by factors of 2 to 5. But also many families never get out from under the burden of lousy local economic situations, low earnings potentials, financial catch-22s, and situations like where small financial setback amplify massively (say, a car breaks down, there’s no money to repair it, so you lose a job, and then get kicked out of your apartment, and then it’s hard to get a new job… etc)

• jim says:

Daniel Said:

“Median household net worth for black families is somewhere in the \$5k to 15k range. So there it’s clear the median family is impoverished…For white families it’s like \$150k. But even \$150k if your mortgage payment is \$2k is only 6.25 years of payments. “

I agree people don’t save enough. But no doubt lots of people have low savings due to ignorance about how to save/invest – the difference between 1% on savings and 10% on stocks is like **huge** over ten years.

Lots to unpack there.

a) what people save, their net worth, housing costs etc are *strongly* geographically dependent. Real Estate values vary by well over an order of magnitude across the country. Black folks are overwhelmingly concentrated in the south where property values are low. I don’t doubt that they’re less well off than whites, but that correction will be significant.

b) How much a person or family has saved is also strongly age dependent, with savings probably peaking just before retirement and declining thereafter.

c) How much people need in retirement is strongly dependent on health care costs, and this can vary with no racial or geographic dependence – enlisted military people from a certain period get a full ride, which can be worth tens of thousands of dollars a year.

d) I couldn’t begin to guess how long-term care insurance, pensions, annuities etc would be counted in “savings” and how much might be provided by governments in different parts of the country.

• jim says:

The difference between a savings account (1%) and the long term stock market average (10%) after X years is:

5yrs 20%
10yr 52%
20yr 160%
40yr 805%

So if you spend your life filling bank accounts you’re going to stay poor.

• Carlos Ungil says:

> The difference between a savings account (1%) and the long term stock market average (10%)

Comparing long-term stock market returns [*] to interes rates one should take into account that short-term interest rates were well above 1% for the most part of that long term.

[*] Annualized long-term returns estimates are usually lower than 10%. If you really mean the average of annual returns keep in midn the “volatility drag”: if the market had cycles of 4 years +20%, +20%, +20%, -20% the average return would be 10% but the actual return would be 8.4% p.a.

• jim says:

Carlos said:

“short-term interest rates were well above 1% for the most part of that long term.”

Daniel referred to the savings of the average American, which is mostly in banks or at bank rates, so that’s what I was comparing to in this case. Sosince 1996 at any rate, 1% is the number. Sure back in the 70s and 80s bank rates were higher.

The value of everything you own is subject to large risk from government policy. If the government f’s up monetary policy or creates adverse economic incentives, your nest egg could be destroyed. There’s no avoiding that with any kind of investment or savings. It’s amazing is how much that’s taken for granted.

• Carlos Ungil says:

> So since 1996 at any rate, 1% is the number.

Over the not-so-long term performance calculations are quite sensitive to the starting and ending points. 24-year annualized return for the S&P 500 now is 9% but 20-year annualized return 6 months ago was 5%. (Better than short-term interest rates, no doubt about that. Bonds would also have performed better than cash.)

https://imgur.com/a/h0Yr4nG

• jim says:

“24-year annualized return for the S&P 500 now is 9% but 20-year annualized return 6 months ago was 5%.”

Fair enough but it was around 5% for what a few weeks? :) Most people can ride that out. If you choose your starting points as peaks and your ending points as crises, surely you won’t do well.

“That chart is for 10yr bonds, long-term bonds are still doing better than equity twenty years later. (Disclaimer: Past performance etc.)”

Yes again if you pick points to start at that favor a particular investment, you’ll do great. If you bought AAPL in July 1999, you’d have earned 300x your original investment. If you know the future, you always do better! :)

• Martha (Smith) says:

Jim said,
“You know when I hear Krugman say “well the stock market isn’t the economy, millions of Americans don’t own stocks!” I cringe. It’s not a good thing that they don’t own stocks. It’s a bad thing. It means either a) they have no excess income to invest, or b) they’re financial ignorant and putting their savings in banks. Lots of immigrants are in the (b) camp, and that’s really sad because they work so hard for so little and save so much, but they earn nothing on their savings so in the end they have to go back to their home countries to retire, unless their kids do well.”

I can’t help a “back in the good old days” anecdote: I once had a four year certificate of deposit that paid 12% interest. I wonder how that would compare to current gains from stocks?

• jim says:

Marth Said: “I once had a four year certificate of deposit that paid 12% interest.”

Let’s see if these Bing links work.

Ha, yep, and savings accounts were paying 10%, I remember that! That was during the big US inflation episode in the 70s to early 80s. Here you can see that 10% was far behind the peak rate of inflation during that period, so while it was better than stocks, putting money in the bank was still a losing cause.

But Paul Volcker killed inflation, ending the “lost decade”, and interest rates collapsed. The market recovered in spectacular style, gaining around 15% annually from 1984-2000.

It’s true that there are periods where the US market hasn’t delivered. People have to make some adjustments when government policy is bad.

However, this chart shows that international diversification probably would have countered the 1970s US stock market, all the more so considering that the US dollar was falling at the time.

3. Peter Dorman says:

When teaching statistics, I used to have an exercise called “What’s the right denominator?” I would divide the students into teams, give them the numerator data (a few vignettes), and their task was to come up with different possible denominators — and the questions those denominators would help illuminate. Then we’d compare what the teams came up with. That was always an “aha” experience.

4. Michael Nelson says:

I like to mix the fun stuff with absurdity and horror. Did you know that permutation tests were first widely used for serious investigation of ESP (“What’s the probability of the subject getting that card right by chance?”), and later in the 1930’s, for “proving” that racial and class stereotypes are accurate (“What’s the probability of the subject correctly identifying the essays written by Jewish authors by chance?”). And you thought Milgram and Zimbardo were unethical! And then there’s the time Fisher made that poor astronomer apologize for saying variance isn’t the best measure of variability, in a footnote to Fisher’s paper, where he also thanked Fisher for the opportunity.

Seriously, though, I find putting methods in the context of their development and evolution, and more broadly in the historical context, can make for many engaging stories.

5. AlexGamma says:

I’m reading Regression and Other Stories right now and I’m loving it!