Monday, October 20, 2014

Survival probability calculator

It’s calculator time again!  When I posted the last one, someone referred to it as depressing.  I hadn’t seen in that way, though I believe this was someone who pays New Jersey property taxes, so I guess any reminder of how much that costs would be a downer.

Anyway, probably everyone will agree that this one is depressing.  In fact, whereas I linked to the other one at the top, middle and bottom of the accompanying post, I’m going to save this one for the end of the post so people can decide whether or not to even use it.

So with that inspiring introduction, I present… a multi-person survival probability calculator!

The immediate idea came from a post at the Hull Financial Planning blog (I don’t follow it, though I’ve followed links to it several times and it seems good) about evaluating your need for life insurance.  Sensibly enough, that calculation requires knowing approximately what your chances of actually filing a claim are.  But also, I’ve long had a bookmark to the SSA’s Actuarial Life Tables and occasionally spent some time studying what they have to say.

What the calculator does is simple: given any number of people, with age and sex, and a number of years, it uses the SSA table to calculate each person’s cumulative probability of surviving for that long, plus the combined probability of everyone in the group making it.

What to do with this information?  

Here are some possibilities:

Ignore it

Don’t even run the calculator.  Sufficient unto the day is the evil thereof, right?  Let’s just go with the working hypothesis that we and our loved ones are immortal.

Figure out why it doesn’t apply to you

Plenty of fertile ground here.  For one thing, are you in better shape than the average of your peers?  And some of the people who die every year start the year knowing they have a potentially fatal condition.  So if that doesn’t apply to you, that must help the odds at least a bit.  Actually, you can do a little better than mere speculation on this one.  I recently came across a pair of web sites that ask you a bunch of questions then predict your personal life expectancy.  They’re called Living to 100 and Blue Zones.  (Note: unfortunately, both require an email address. But I think it might work to use a fake one.)

And then there’s another sobering statistic: for some age groups, suicide is the second leading cause of death.  That seems like one where you can have at least some idea of how big a risk factor it is for you personally.

Examine your priorities

The question, as I see it, is “If the moment comes when I discover that I’m going to get the bad side of this equation, what will I think about my choices?  Is there anything I could be doing to make that hypothetical future self feel better about my life, without messing things up for the other future self who’s living in the much more likely hypothetical world where everything turns out OK?”

Freak out

Plenty of directions to take this--extreme risk-aversion, frantically piling up experiences, depression, …  I’m sure there are others, and you could also bounce between them all.

Acceptance [1]

Maybe your priorities are in decent shape and you’re willing and able to take this in and see it as just one of the many uncertainties of life (“Anything can happen”).

I think I’ve done some of all of these.  Even the first one, since I spent a few days thinking “Yeah, I probably don’t need to do that” between thinking of the idea and figuring out my own number.  I haven’t freaked out much, but there have certainly been moments of internal panic.  I wasn’t expecting a double-digit number for my 20-year downside risk.

And finally... the calculator! [2]




[1] I realized partway through thinking about and writing down these responses that they map fairly well to the “stages of grief”, if you put anger and depression together under “freak out” and group “examine your priorities” in with acceptance (which makes sense, as a different form of acceptance, i.e. “This is real and it applies to me. What am I going to do about it?”).

[2] Note re data/privacy: this calculator operates entirely in your browser and doesn’t transmit anything.  I kind of wish I had made it at least tell me when it’s been clicked, because I’m curious how many people will actually run it, but I didn’t.

Wednesday, September 10, 2014

The Do-It-Yourself Decision Calculus: Cheaper, Better, or Fun

File this under rules of thumb or guidelines that I formulated in my head at some point and refer to occasionally.  This one is probably 5-8 years old.

The rule: it's worth doing something yourself (vs. buying a product or hiring out a task) when it's some compelling combination of cheaper, better, or just plain fun to do.

A particular decision could be made on small contributions from each factor, or it could be based entirely on the strength of one of them, but it has to meet that test somehow.  (If you flip it around, it sounds pretty obvious: if you'd be suffering to do/make something that will cost more and turn out worse, obviously that's something you should buy or do without.)

My primary examples: beer and applesauce.

For example, that's Two Hearted on the right.

Beer


I've been asked before whether I brew beer, because it seems like something I'd be into.  But in fact I've never even really been tempted.  It certainly doesn't sound very fun—a bit of planning and cooking, a lot of waiting, and a lot of tedious sterilizing and processing—and while home-brewed beer is cheaper, my impression is that it's not way cheaper, and at the rate I drink beer (which I guess might increase a bit if I had larger quantities sitting around, but not drastically) it would take a long time to make up for the investment in gear.  But the biggest factor for me in this case is the "better".  There's so much really good beer out there, I just don't see myself managing to make something that would compete.

Applesauce

You can buy applesauce cheaply and you can buy pretty good applesauce (not cheaply), but can you get applesauce that's as good as our home-canned Fuji/Honeycrisp blend?  I'm not sure you can.  And can you get it for about $1.70 per quart?  No, you cannot.  I can't say it's particularly fun (as a small-scale activity with friends or family it might be, but the way we do it these days is I stay up late processing a bushel at a time with a hand-powered Foley mill), but it's totally worth it.

Other cases:

  • Knitting: it's not even close on price (decent yarn is expensive, manufactured goods with the same quality materials cost less), and I think in most cases there's no quality advantage.  So this is something you should do if you find it fun (of course the fun can be in wearing or giving away something you made, but it's probably best if you enjoy the activity itself at least somewhat).  I used to, but at some point it stopped being worth it for me.
  • Home/car/bike repairs: all about cost.  Though in some cases I think quality benefits, because you have the time and motivation to be as careful and do as good a job as you can, whereas someone else's main goal might be to finish and get paid as fast as possible.  Obviously quality can also suffer if you don't quite have the hang of what you're trying to do (solution: more YouTube videos!).  There can be a fun factor, too, if we stretch the definition of "fun" to include "sense of satisfaction at one's own increased competence and self-reliance."
  • Roasting coffee: I would like to get back to doing this.  The cost is roughly even, but I found it enjoyable.  And while there's no shortage of good coffee out there, I think in this case there's something to the claims that extreme freshness makes a noticeable difference.
  • Granola, hummus, yogurt: I've been making these three recently (yogurt very recently, and it might not stick, but the other two are part of my routine now).  I do like my versions, but there are plenty of store-bought versions that are excellent.  So this is purely a cost issue.  Not that we're actually saving money, I don't think, but we get to eat more granola and a lot more hummus than we would if we were paying store prices for them.


Saturday, September 6, 2014

A listicle: Dead things encountered on the bike path (plus some meta-blogging)

I'm guessing nobody had marked their calendar and was getting impatient, and I certainly haven't gotten any emails about it, but I'm behind schedule.  Which would be more understandable if I had set myself an ambitious schedule, like something tied to a day of the week.  In fact what I have is "monthly" (meaning "each post should be no more than 30 days after the last"), and I've still blown it.  I have an excuse, though, sort of: I really want to get the promised* theodicy post done, but so far I've found it daunting to even try, and difficult to carry out. That is, I've thought about it a lot, drafted a little, and not gotten all that far.

So in the meantime, some lighter fare...

Dead Things I've Encountered On The Bike Path


1. A fish
I'm pretty sure this was the first, and it might be the weirdest (or it might not. I kind of find #5 weirder).  When it rains enough, the Wissahickon Creek gets excited and rises fast, which means some sections of the bike path flood fairly easily.  One of the first times this happened after I started commuting, I made the mistake of hitting the first flooded section and thinking "well, I can pretty much lift my feet and coast through this."  A few more slightly larger water hazards later, I had gone too far to turn back and ended up pedaling through about 18 inches of muddy water for a quarter mile (and also running off the path at one point because I couldn't remember on which side of a particular tree it went).

That's all for illustration, because I think that time I learned my lesson and took another route the next day.  But after one of the first floods after I started commuting, on the way to work I was surprised to see a small (4- or 5-inch) fish in the middle of the path.

2. Goslings
Not at all surprising, really, since there are a lot of geese that live along the path year round, and in the spring they all get super pissy and aggressive in defense of their nests and then their cute little hatchlings.  Though the normal attitude of a bike commuter to geese is antipathy (due to (a) the aforementioned aggression, (b) during other seasons, the aimless wandering across the path, and (c) all the poop), it's still sad to see dead goslings.  But it happens, I assume from collisions with bikes.  I've seen maybe four total.

3. Mice
I've seen one or two before, and I just saw one the other day.  I feel like I've had more close calls with squirrels and chipmunks, but it's mice that I've actually seen dead on the path.  Maybe other rodents are sturdy enough to survive a hit from a bike, or at least run a ways before succumbing to their injuries.

4. A fox
Not actually on the path, but right next to it.  It was lying curled up in a fairly normal-looking pose, so I don't know whether it was hit by a bike or not.  I assumed not, actually, though I didn't have any good ideas about why it would have chosen that spot to die from something else. 

5. A deer
This was before I had a smartphone, or else I'd have a record of this one.  If I'm remembering right, it was a young buck with the beginnings of some antlers.  In any case, it was definitely a deer, looking like it had gone down face first, with a bit of blood coming from its mouth.  It was right below the regional rail bridge over the beginning of the trail along Lincoln Drive, so my best guess is that it was up on the tracks and either got actually hit by a train or was surprised by one with nowhere to go and ended up falling on its head down the pretty steep cliff.



6. Trees
So many downed trees, in all shapes and sizes.  Usually you have to climb over, occasionally under, and sometimes it's the top that falls across the path so you end up having to pick your way through among the leaves and branches.  Once there was a really big one that fell away from the path but was growing so close to it that it took a bug chunk out of the side of the path, leaving a sheer drop into an 8-foot hole.

7. A Pontiac
Ok, so cars aren't alive.  But it still works to talk about them as dead. Such as when they've been jammed beneath two bridge supports that are at least a foot too narrow.


* Sort of promised here, and actually promised in person to a friend.

Tuesday, July 29, 2014

Cobain's Sarcoma

Why does the idea of heroin addiction affect me so strongly?  I don’t really know.  I mean, there are people in my family with cancer, too, but somehow that doesn’t get to me in the same way.

Possibly it’s because I have so much invested, philosophically and emotionally, in the idea of freedom, and heroin goes in and attacks the will like almost nothing else.  I tried a while ago to figure out whether “cancer of the will” was a phrase that people use to describe addiction.  It seemed like the answer was “yes, but not widely”, so I don’t know if I heard/read it or came up with it myself.  But it seems apt to me.  Heroin addiction isn’t an external force overpowering the will, but a malignancy of the will itself.  How do you fight something like that?

Fuck heroin.  It makes me so sad and so angry.


And now here are a few songs about heroin, some of which sometimes make me cry…





Friday, June 27, 2014

Annals of obsession: Bike commuting

In which, instead of writing another "first in a series" post, I actually add to a series! Though full disclosure: I outlined all the topics connected to cycling and bike commuting that I thought I might have something to say about, and the list was long. So there might be another series coming. But for now, I'm here to talk about the period during which bike commuting was a major obsession.

Getting started

Having lived in Mt. Airy and worked in Center City the whole time I've lived in Philly, I basically always wanted to be riding to work but never thought it was possible. It's about 11 miles each way, which seemed like too much. And that's not to mention the sweatiness issue.

I especially felt like I should have been able to make it work after starting my current job, with its nonexistent dress code and easy flexibility, but I never did. I thought about riding one way and taking transit the other, alternating direction each day, to make it not so hard, but I never did that. So what finally broke the pattern and got me to start? A coworker did it. A friend from work who lived nearby started, and behold, it turned out it could be done! The secret was to do it. We both started out riding a couple days a week and worked our way up, but it turned out the round trip wasn't as impossible as I had imagined it.

Getting hooked

Having finally gotten past the barrier of getting started, I dove in. Some items to illustrate the situation:
  • My first commute was on August 13, 2007. Some time in September, when I was still riding only 2 or 3 times per week, I started a spreadsheet to track my mileage and keep notes (the first entry was in honor of my first flat, after 396 total miles).
  • I set myself a rule that I had to do the ride the first time I faced some new set of conditions (e.g. heavy rain, cold, high winds). That way I would know what it was like and could choose to skip it the next time, but I couldn't be daunted by imagined hardship. This rule served me well, except on the first day it snowed. I went down twice, banging my knee pretty hard on the second one.
  • I geared up immediately. By the end of 2007, I had a variety of wacky synthetic clothing items, several lights, a few new tools, and neoprene shoe covers. (As mentioned in the previous bullet, snow and ice were a problem that first winter. So in late 2008, I got studded tires.)
  • I joined BikeForums.net on August 26, 2007, (i.e. less than two weeks after my first ride) and posted for the first time on September 13. Though the spreadsheet might be considered more extreme behavior, BikeForums was the center of what made this a major obsession. It's possible I kept up with the commuting forum for more than a year. Anyone who has spent significant time in active online forums will appreciate how shocking that is.
  • The spreadsheet started out fairly tame, but grew quickly to what I think would be considered a "maybe a little unbalanced" level of detail. Actually, I think it gets its own heading.

The Spreadsheet

At the beginning of 2008, I started taking notes on every ride (ride time and distance, weather conditions). It also tracks cumulative commuting and recreational mileage, skipped rides, maintenance, and spending.

Some factoids I know thanks to the spreadsheet, to illustrate how ridiculous it truly is:
  • As of today I've ridden 25,744 commuting and 5,790 recreational miles since my first commute.
  • Coldest ride (tie): 12 °F, February 11, 2008, and January 28, 2014.
  • Hottest ride: 102 °F, July 22, 2011.
  • Windiest ride: W 29 mph, gusting to 46, February 12, 2009. 
  • Total spending: $5,089.45. The most recent item was a new studded rear tire, since the old one lost its ability to stay on a rim. According to my somewhat-convoluted formula for calculating how much I would have spent commuting by other means, I'm only $105.72 away from breaking even (at last!). Unfortunately I need a new cassette at the moment, so that'll set me back a couple weeks.
  • In mid-August of last year, my current bike (pictured above. It's my ideal commuting bike, bought piecemeal and built in March 2011, to replace the circa-1990 Sirrus Sport that wouldn't take fenders) became my highest-mileage bike. It's now a month or so away from hitting 10,000.
  • The data I have isn't fantastic for the Wolverine's mileage, but she's right around 500 lifetime miles, mostly commuting (i.e. to and from daycare).
Ok, so that makes it sound like I'm still dangerously obsessed, since a lot of those items are recent, and clearly I'm still maintaining the spreadsheet.  However...

Calmer times

I still love commuting by bike. And I still love data. But I haven't read BikeForums in a few years, and I pretty much never end up down the rabbit hole on some question of gear or technique the way I used to do all the time. I have my habits and I maintain them, but they don't take up much time or attention. I figure the obsessive phase of this hobby wound down some time in 2009 (replaced by: endurance cycling! Not such a radical shift. In fact they overlapped. More on that when this series continues).

And finally, a note on what I like best about bike commuting. Is it the view of the Schuylkill instead of the one from the bus (mostly urban decay) and subway (darkness)? Is it being in shape, or the 20 pounds I lost pretty quickly after starting? Is it the double tailwind days (so rare!) or the quietly falling snow? Those are getting closer. It's the joy.


Saturday, May 24, 2014

Super-fancy and overly complicated house cost calculator

This tool originated from a conversation about the idea of buying a new house and living there for five years. My reaction to the idea was mild (or maybe moderate) horror at the enormous transaction cost. It's easy to do the math on that part, but it got me wondering what the actual all-in cost of such a decision would be. Which meant I needed to figure out the all-in cost of buying, owning, and then selling a new house vs. the cost of continuing to own the one we have.

This house looks nice. Maybe I'll buy continue to own it!
So I made a fairly complicated Google spreadsheet to figure it out. (The answer I got was suitably striking: $50,000. But I've since discovered several bugs in that original implementation that caused it to overstate the difference.) Then I wanted to share my beautiful creation, but linking to a Google spreadsheet seemed dumpy. A javascript calculator would be much nicer. Too bad I didn't know javascript. Fast forward a while to when I decided it was time to learn javascript, and I knew just the project to use for practice.

The idea of this calculator is to figure the total cost of owning a house—mortgage interest, property tax, insurance, and, importantly, the opportunity costs of having a bunch of money tied up in home equity.  When I say it includes insurance, what I actually mean is that it lets you enter a fixed amount and/or a fixed percentage of home value to add in, so it doesn't do anything very smart to figure out insurance cost, but it does provide a place to account for it. Also utilities, if you want to try to estimate them. Different houses will definitely result in different utility bills, but it's hard to predict what the actual relationship will be.

The calculator is on a page of its own here. You can go there now, or stick around for what's probably a lot more explanation and comment about the assumptions, parameters, and output values than is really necessary or useful.

Assumptions


Inflation: One of the biggest pitfalls with long-term calculations like this is dealing with inflation. For example, it would be easy to fall into the trap of thinking that you made a nice profit on a house if, say, you bought it for $150k and sold it ten years later for $200k.  Actually, if inflation averaged 3% over that time, your real return would have been -$2,403. Before transaction costs. So it's really important to account for inflation in these types of models. The calculator attempts to provide all values in current dollars. So, for instance, after a while the sum of "Principal paid" and "Remaining principal" won't add up to the purchase price, because both will have been discounted for inflation.

Real opportunity cost, real appreciation rate: This is a continuation of the above, but I wanted to emphasize it. The opportunity cost rate is meant to be a real (i.e. inflation-adjusted) rate. So if you think you can average a nominal 7% return on your investments but that inflation will be 3%, your opportunity cost rate should be 4%. And the appreciation rate should be not how much you think the house's value will grow, but how much it will grow above the rate of inflation. General rule of thumb: it won't. The long-term average behavior of house prices is that they keep up with but don't outpace inflation.

Selling: Since I was focused on a "buy then sell" vs. "do nothing" scenario, I set this up so that it requires a sale date and the overall bottom line is post-transaction-costs. I don't think that's necessarily a bad thing, since transaction costs are ginormous and it's easy to forget about them when doing rough estimation. But if you want to do a run for "keep indefinitely" you can put in a big number for "Years before selling" and set the "Selling cost" to zero.

Tricks


Hiding columns: There is an excess of columns in the output table. If you click on a column header, that column will disappear (and you'll get a button to put it back if you want).  Your choice of what columns to show and hide should be persistent (unless you reject the cookie. Or there's a bug).

Calculating the status quo: To figure out the current and future costs of your current situation, just set the purchase price to what you think the current value of your house is, the down payment percentage to the number that makes "Purchase price" minus "Down payment amount" equal to your current remaining mortgage principal, and the mortgage term to the amount of time left on your mortgage (use decimal years if needed).

Saving your input values: The results page includes a link back to the calculator with the values you chose encoded in the URL, so that they'll override the defaults. I.e. it's suitable for bookmarking or for opening multiple copies of to try different variations.

Input parameters


There's really no point in commenting on all of them, is there? But here are some notes.

Purchase price: Starting off easy.  It's exactly what it sounds like.  Though it's treated as the value of the house, so if there are shenanigans like a "seller assist" involved, it would probably make sense to back those out and adjust the down payment to make the initial mortgage amount match.

Down payment amount: Not editable or used directly, but it updates based on purchase price and down payment percentage to show the actual dollar amount that they imply.

Mortgage rate: The built-in default is 4%, which is in the ballpark for a 30-year fixed loan at the moment. The default is the most recent 30-year fixed rate from Freddie Mac's Primary Mortgage Market Survey (loaded from Quandl. Do people know about Quandl? I just found out about it. It is awesome).

Years before selling: As mentioned above, you can ignore the selling aspect by setting this high and "Selling cost" to zero.

Show every X months: This controls how many rows the output table will have. It's not smart enough to always show the last month if it's not divisible by this number, but fortunately multiples of 12 are very divisible numbers.

Real opportunity cost rate: This should be the average inflation-adjusted return you expect from money you have invested. I.e. if the money tied up in the house were in a retirement or investment account instead, how much would you expect it to earn above inflation. The default, 5%, is on the conservative side of fair for a long-time-horizon diversified portfolio. At least historically. If you think the future is not so bright, adjust this down. But don't succumb to knee-jerk pessimism, either.

Note that this assumes that if the money you have invested in a house were available it would be invested in a diversified portfolio. If you believe yourself to be the sort of person who benefits from the forced savings aspect of having a mortgage, and who wouldn't manage to keep that money invested if it were in a more liquid form, you should adjust your opportunity cost rate down. In the extreme example, I guess if you thought you would spend every penny that didn't go toward the mortgage, you could make the opportunity cost rate zero. Or even negative. I would need to give more thought to whether that would produce sensible results.

Inflation rate: There are a lot of different ways this could go. The average since the mid-80s is something like 2.8%, but since we don't seem like we're that close to getting out from under the Great Recession, I made the default 2.5%. That's leaving aside the secular stagnation hypothesis, which could mean it'll be much lower or could mean—if the idea gains wide acceptance and the Fed acts to counteract it—that it it'll be significantly higher. So yeah, 2.5% seems a decent guess to me.

Buying cost: As I understand it, sellers and buyers usually split transfer taxes. In Philadelphia, those are 4%, so I estimated 1% for mortgage fees, title insurance, etc., and made it default to 3%.

Selling cost: Defaults to 8%. Hopefully it's lower for some people, but in Philadelphia it's probably slightly higher, given the traditional 6% agent commission and the seller's half of the 4% transfer tax.

Real appreciation rate: As noted above, the long term average is for houses to keep up with inflation. Obviously the short-term behavior can deviate a lot from that. But I don't know of a reason to predict something different happening in the future.

Property tax rate: Philly now uses actual value with a homestead exclusion. If your taxes are based on some other calculation, you'll have to do some math. It should be possible to use this and the "Property tax exclusion amount" field to get to the right answer, though.

Property tax exclusion amount: Subtracted off of the current home value before it's multiplied by the property tax rate.

Misc expenses (% of home value): What it sounds like. For each month, it's multiplied by the appreciated home value for that month.

Misc expenses (dollar amount): Unlike most of the amounts in the table, this is assumed to increase with inflation, so will not be discounted by the inflation rate.

Reuse results window?: If this is unchecked, it will pop up a new window every time you hit the button. If checked, it will make a window and then reuse that window for subsequent runs.

Results table columns


Wow, this is getting long. Hopefully people bailed out and followed the link above.  Here it is again if you hung in this far but have had enough.

But for those who are ready to stay with me to the bitter end, because there's something they're procrastinating that they really don't want to go do, I shall press on with a few notes on the output values.

Payment: This is the amount your mortgage company will charge you for principal and interest, based on the amortization calculation. It might seem weird that this amount changes every month for a fixed-rate loan, but that's because the bank is charging you based on the nominal balance of your loan but the table is showing current dollars. Every month the amount you have to pay is actually, adjusted for inflation, a little lower.

Interest: The portion of the month's payment that goes to interest.

Total principal paid, Remaining principal: These move in the direction you would expect, but the total paid grows a little slower than you would expect and the remaining principal shrinks a little faster, again due to inflation.

Current value, Equity: Current value is the purchase price times the compounded real rate of return. Which, by default, is zero, so this column is pretty boring. Equity is that value minus the remaining principal. When you've borrowed money to buy an appreciating asset, inflation makes you slightly richer.

Opportunity cost: This is how much you're losing during the month due to a) the money that's tied up in the house not being invested more productively and b) the money you've spent on interest, taxes, and opportunity cost in prior months being gone instead of being invested and earning money for you.

Monthly bill: Not directly relevant to cost, but since I had the quantities available I figured it would be nice to see the actual monthly cash flow implications. This is your monthly principal and interest payment plus monthly misc expenses and property tax.

Monthly all costs: This is what it actually costs you to own the house for the month. Compared to "Monthly bill" it adds opportunity cost but subtracts the amount that goes to principal.

Totals: Then we have some running totals of the quantities described above. Discounted for inflation, of course.

Sales cost per month: The last few columns are concerned with the effects of selling. This is the total cost to sell the house in the given month divided by the number of months since you bought it.  I.e. if you sold this month, how much would you have paid per month just in transaction costs.

Total appreciation: This is equity minus total principal paid.

Note: My default scenario has home value matching but not beating inflation. So if you're not making money (in real terms) on the house, where is all this appreciation coming from? One possibility is that it's a bug. But I think it's not—I think it works out that way because the amount you actually paid for the house is constantly shrinking due to inflation, but that doesn't have an effect on your monthly costs. You only profit from that difference when you sell. Anyone who can actually explain this so it makes sense (either an understandable way of saying why the calculator is right or an explanation of why it's wrong), please comment.

Final monthly cost: The sum of "Total all costs" and total sales cost minus total appreciation, divided by months of ownership.

Final total cost: The total amount the whole thing cost you, post-sale.

Fin


At last, the end!  If you haven't already, it's time to go to the calculator!

Postscript: The New York Times and I have so much in common


I wrote this post (all but a couple paragraphs) two nights ago, on May 21. The code wasn't quite ready, but I hoped to have everything done and posted in a day or three. Then what should I find in my Facebook feed the next morning? This! A new tool by the New York Times's "The Upshot" blog that's remarkably similar to mine, except for being way nicer looking, more usable, and generally great.

I must admit to having felt some chagrin. It's true that I'm not in competition with the Times. And what we've both done is presented a solution to a particular math problem that affects lots of people's lives, so you can't really consider that creative work that you would expect to be unique. But still, I built this tool because, until yesterday, I hadn't seen anything like it. So it took the wind out of my sails a little to have this thing I've been thinking about since last July (that's when I made the spreadsheet. The javascript part is more recent) so precisely clobbered.

The one revision I've made to my tool based on looking at NYT's is that I had put all the transaction costs on the selling side, not realizing that transfer taxes are usually split and thinking the other fees are small enough to ignore, when in fact they can add up to real money.

I would say the one thing I like better about my tool is that the table lets you see intermediate amounts and makes it a bit easier to see how things add up and fit together. I also think their opportunity cost default is too low (though some might argue that I've gone too far in the other direction), and while it's cool that they try to account for taxes, I suspect they treat all property tax and mortgage interest as deductible against the entered marginal rate, and I think in a lot of cases—maybe more often than not—that would overestimate people's tax savings. And I provide a bookmarkable link. I don't see a way to save your numbers in their tool.

On the other hand... there are too many things to mention. But I will say that the comparison to renting, which is somewhat central to the Time's tool and totally absent from mine, was not a priority for me because I don't feel like I could find a suitable rental in my neighborhood at any price. In New York there's a pretty diverse and comprehensive rental market, but I think in a lot of places, there just aren't good rental options to be weighed against buying.

So that's the story about that. My little tool might be 99% superfluous now, but I like it all the same.

Thursday, April 24, 2014

A personal finance spectrum

There was a lunchtime presentation at my work the other day on "Financial Wellness", and one of the things I took from it is that it's really hard to give a presentation like that that's pitched to what your audience needs.  Because there's a good chance the audience will include a range of people in vastly different financial situations, some of whose needs, in terms of advice and education, basically don't overlap at all.

So that got me thinking about what I would do if I wanted to give such a presentation, and I figured I would have to be ready with at least three different presentations, then start off with an (anonymous) poll and deliver the one that was pitched to the majority of the audience.

So here's what the poll might be like.  So as to avoid anything too value-laden, I figured the scale should be something that's not strongly suggestive of a ranking.  Though I might not have succeeded.  This is certainly reminiscent of the old Terrorism Threat Level and the forest fire risk scales.  Not entirely value-neutral.  But actually I kind of like that the red end is suggestive of danger.

  • Spending consistently exceeds income. Large and growing debt.
  • Spending is sometimes more and never less than income, so debt is either growing or flat.
  • Spending sometimes exceeds income and debt expands, but when income ends up higher at least part of it goes to reduce the debt.
  • Spending equals income, because it expands to absorb whatever comes in. Financial decisions are based on immediate cash flow and there is no expectation of increasing net worth.
  • Spending roughly equals income, but when income is higher much of the difference is kept.
  • Income slightly exceeds spending. Able to save, but mostly saves up for things rather than accumulating very much.
  • Income consistently exceeds spending. Saves a small amount (e.g. enough 401k to get matching, plus a little in savings).
  • Saves 10-15% of net income. Probably on track to retire at Full Retirement Age. Spending is correlated with income, but the coefficient is below 1.
  • Saves 10-15% of net income, but spending is not strongly correlated with income, so additional income or reduced expenses will have a significant impact on the bottom line.
  • Saves 40% or more of net income. On track to be financially independent before Full Retirement Age.
  • Saves 60% or more of net income. On track to either retire soon or end up quite wealthy.

This particular spectrum is very focused on income/expense balance.  I figure that correlates with factors like spending habits, debt use and payment habits, approach to high-impact financial decisions (like how much to spend on housing and transportation), and level of confidence in managing finances and investing, but I'm sure you could put those other factors on the same scale and plenty of people would find they're best described by a range of different hues.

And now... the poll!  Just for fun.  Fully anonymous (I think. I certainly have no idea how I would track clicks back to people. At most an IP might be logged in a database on the server of the free poll creation web site I googled up, where nobody would know or care what it was about).

Where do you fall on the spectrum?
  
pollcode.com free polls