Hacker News new | ask | show | jobs
by toado85 3761 days ago
Hey guys... I made the page. Just saw lots of traffic coming from here so figured I'd come check it out.

A couple things I figured I should address after reading the comments:

1) Yahoo!'s historic S&P 500 data does not factor in dividends. So the returns would likely be 1-2% higher each year (which over time makes a very big difference). I should probably add a note on the page mentioning this.

Here was my conundrum when making the tool: I picked the S&P 500 because it's the only index that allowed me to pull very, very old data (nearly 70 years) using Yahoo! Finance; plus, it's often the "go-to" index for discussing overall market performance. But it's not "real" in the sense that you can't actually buy shares, and it doesn't pay dividends. So I could make up my own method for factoring in dividends, but I wanted to go strictly by the numbers. When you factor in financial advisor fees / bad decisions that new investors make, it's probably enough to "counter" the lack of dividends, if you want to look at it that way.

Plus, sites / companies are notorious for over-stating how much you can get annually by investing. I'd prefer to under-state it, if anything. Don't want to sell false hopes.

2) Regarding incremental, small deposits (and potential transation fees)... it's actually very easy to set up auto-investments in index funds that match the S&P 500 without ever incurring any fees. You could do $31 on the first of every month and basically simulate this.

3) Inflation would be useful to factor in, but it would also add confusion. This could be a cool add-on, but I'd have to think about the clearest way to demonstrate it. So would the 1950 daily amount be equivalent of $1 today? (so I'm guessing 20 cents or so?)

Hope you guys enjoy the site. Feedback is great (positive or negative - I'm not sensitive).

9 comments

Nice site.

You may want to look at Robert Schiller's (of Case-Schiller index fame, among other things) data set available here: http://www.econ.yale.edu/~shiller/data.htm He has S&P dividends by quarter back to 1871.

Fees do make a difference, though they are pretty small in e.g. VFINX -> VFIAX. If you want to be super realistic you could build up the dollars a day at say the federal funds rate until you hit the VFINX minimum, then accumulate $100 batches at that rate before investing them in the S&P 500. Once the account hits VFIAX switch to that for fees. There's also taxes but that's just a mess because there are so many possible scenarios. Probably this whole paragraph is just overkill.

As for as inflation goes, the big thing is not the current amount of money, which after all is today's dollars and so easy to understand, but to try to get across the fact that in 1950 $1 was a heck of a lot more than it was today (about $10). Still not a huge amount of money perhaps, but not something you would just drop in a tip jar either.

That also gets to the point that is kind of a crazy investment strategy. In real terms you are investing the most when you are least able to afford to. Since it isn't supposed to be a serious strategy, instead more of something to think about that's no big deal, but it's worth noting.

Just updated to include dividends / dividend-reinvestment on by default. Thanks!
Regarding inflation, you can use the the CPI (http://www.bls.gov/cpi/), specifically the CPI-U (all urban consumers), not seasonally adjusted. For example, $1 in 2015 was $0.10 in 1950. The BLS actually has a little calculator you can play with (http://data.bls.gov/cgi-bin/cpicalc.pl).

It's fairly straightforward, but if you want a script to start you off, here's a little something I wrote in perl for my own fun: https://github.com/Amorymeltzer/sandbox/blob/master/inflatio...

I think it is a great visualization and it helps to illustrate why saving and investing should be part of everyone's strategy to have something set aside when they are older.

And I think it illustrates that it doesn't have to be a "lot" of money to make a big difference over time.

It's also rather misleading, as you point out, because you can't invest $1 (or often even $31 a month). I assume you have seen the mr. money mustache (MMM) web site. His approach to communicate the value was to discuss mutual funds which reflected broad diversity in the market. Many of them have quite long histories as well. Of course with mutual funds you have to factor in taxes (which is really hard to do in a general way) since funds transact sales and that results in the realizing of short and long term gains and losses.

Aside from the difficulty of actually investing $1 a month I found the visualizations excellent and the floating bubbles pretty cool too.

RE: 1, this reinvests dividends:

http://finance.yahoo.com/q?s=%5ESP500TR

S&P 500 Total Return Index. Calculated intraday by S&P based on the price changes and reinvested dividends of SPX <INDEX> with a starting date of Jan 4, 1988.

I'd love to use something like this or SPY or VFIAX rather than ^GSPC, but that's 38 fewer years of data available. I'll figure something out, though.
Hey Toado,

Your webpage gave me a screen takeover popup that was unavoidable on my phone :/ Even hitting back button would re-launch the popup.

Chrome on iOS if it helps.

Ewww... never heard of that happening. I tested on iPad with Chrome / Safari / Firefox. Is this on iPhone? If so, which one?
Here's what it looks like on iphone: http://www.webpagetest.org/result/160309_JW_14VA/
Yes, iPhone 5.

I started entering year before popup came up, then wouldn't let get get out and see the actual stuff. Quite annoying.

I had one as well. Safari on iOS 9.2.1
Through Yahoo's api, can't you use "Adj. Close"? That's what I do for historical S&P.
That's technically what I'm using, however, if you look at the S&P 500 historic data - http://finance.yahoo.com/q/hp?s=%5EGSPC+Historical+Prices - the Adj. Close is the same as actual close (meaning there's never dividends or splits to account for). That Adj. close is what the main site - http://stockchoker.com/ - uses for all symbols, so for pretty much everything else, splits and dividends are accounted for... but nothing else I've found goes back anywhere close to 1950 (and is also representative of the stock market as a whole).

So that's why I'm leaning toward still using the S&P500 data, but adding an "average" dividend payout each year (which I'll do some research on.. guessing a little over 1% a year).

On inflation - it might muddle the calculation. Unless you're saying "Invested the equivalent of today's dollar"

I believe that CRSP [0] has historical indices that do include dividends.

[0] http://www.crsp.com/

Found this - http://www.multpl.com/s-p-500-dividend-yield/table - which I'll use to add a "include-dividends" mode (on by default).
You sold me with this simple site. I can handle $31 a month easy. I'll set up a monthly investment into index funds later today.

I have a Sharebuilder account, but would there be a better site to do this from nowadays? Preferably with a good companion app?