Live data from Hacker News

See how a dollar would have grown over the past 94 years [pdf]

newyorklifeinvestments.com

131–139 of 139 posts

Re: See how a dollar would have grown over the past 94 years [pdf]

#131

This isn't a chart of returns; it's a chart of who had nerves of absolute steel. Be honest, who here has actually lived through a major dip and not been tempted to smash that "sell" button?

I don't get tempted to smash the sell button during major dips, I get tempted to smash the buy button. The only time I smashed the sell button was during the stock market bubble in 1999. I wanted to buy during this year's dip but I took a look at my asset allocation and I was still way overweight in US stocks compared to my target so I couldn't justify it. Hopefully people freak out even more next time. Stock market…

> I don't get tempted to smash the sell button during major dips, I get tempted to smash the buy button.

Same. Not quite dollar-cost averaging, but when I'm confident in the long term outlook of a company, if they dip, I like to buy more.

Just before the COVID lockdowns, I sold everything and went to cash for a month or so. Bought back all my previous positions right about the bottom of the dip. Massive gains on that one... hard to time things that well all the time though.

Edit: One additional thing is that you can't be emotionally invested in the investments. That's just a recipe for poorly thought out choices, and ultimately a disaster

Re: See how a dollar would have grown over the past 94 years [pdf]

#132
post #60

Astute observers will note that the $1 in stocks gets margin called shortly after the initial investment in the late 20's and so in a sense the return over 100 years is 0.

Incorrect, the DJIA lost around 90% of its index value in the Great Depression, so the theoretical portfolio in this exercise would have declined to around 10 cents, but holding an asset at a negative return does not result in a margin call if one is not trading with leverage. In practice, it would have been a toss-up between individuals' outcomes because index funds as we are familiar with them today did not exist a…

Crickee you're entirely correct, the negative part of the graph is simply ROI and not value. Thank you for the correction.

Re: See how a dollar would have grown over the past 94 years [pdf]

#133
post #130

Earlier quoted context omitted.

The chart you linked isn't inflation indexed. I believe the chart below is the one you should have been linking (at least one that's public, the reports I get are subscription only so I can't share those): https://fred.stlouisfed.org/series/DFII10 Anything less than 4% for a real return of 2-3% after inflation falls under low interest rates, and as you can see 2003-2022 this period matches that criteria, with real ne…

> The chart you linked isn't inflation indexed. Neither was your comment. The alternative reading "You lose money from the inflation when the rate of interest is below the inflation rate which it almost surely was given the several decades of almost zero low-interest REAL rates in that time period" wouldn't have made much sense. The interest rate is below the inflation rate when the real rates are negative - not just…

> Neither was your comment.

The comment I made refers to the underlying process of inflation indexing a yield curve in loose terms. You seem to be nitpicking on semantics for no real reason.

You also seem to have messed up your math with regards to calculating the loss in purchasing power due to inflation of the principal and interest (coupon) of an investment.

Inflation in 1990 was 5.4%, the return was 8.2% for that year, but I see you rounded that down to 8% in your example, so I'll stick with what you rounded down to in my breakdown below but the real rate of return for that year without your adjustment was 2.8% without taxes. Taxes play an important part in figuring your breakeven.

In your 1990 example, you make $8 as a coupon payment in interest for that year at what appears to be an 8% interest rate. Inflation in that year was 5.4%.

Depending on your ordinary income capital gains bracket/investment you may be taxed on this payment anywhere from 10% to 37%.

That's between $0.80-$2.96 of that $8 in tax.

The real profit/gain after inflation losses was $2.60, absent taxes, but you have to account for your real investment which includes taxes. You are taxed on $8, not that $2.6.

We'll assume its the upper tax rate, which taxes amount to 2.96. $2.60-2.96=-0.36, it is negative so you lost purchasing power on your principal for that year but still made a real return of $5.04 in that years dollars.

Lets look at the next year (its a 10 year bond after all). You get 8%, inflation is at 4.8%, real is $0.30 after subtracting out the tax. You still haven't broken even in purchasing power from the previous year, and if you do the autosum each year you end up making an ever so slight profit because stars align, and only because you caught the top of the bond market.

Over the years rates dropped to 6% down to, 4% in 2001 which enters our low rate era.

Now the general consensus at the time in the 90s was it was a great time to buy bonds, but that was only because the stock market during that time on average lost 6% per annum (iirc), and gold was down too. Nearly everyone was losing some money in purchasing power.

Lets say you reup that bond, $100 in 2001, rates are at 4%, You invest 100 you get 140 at the end, $4 coupon payments per year. Inflation is at 2.8%.

Real gain for that year is $1.20, but you are taxed on $4, which is $1.48 in tax (@37%). Your net profit after taxes: -0.28 in purchasing power.

This gets worse later on when rates go negative as a result of inflation (2011, and 2020), you pay people at a loss for the privilege of loaning them money.

In the long-run you barely break even, maybe, and that bond bubble still hasn't popped yet.

Now these are pretty good estimates considering you don't need to know lot of financial math so long as you handle the calculations properly at the right time, there is some error but not much.

The exact gain amounts also greatly depend on how accurate that CPI calculation is and in most recent years its almost borderline fabrication to the point of uselessness. Shadowstats produces reports on CPI utilizing the older more accurate methods, and those reports show much worse losses which agree with subjective observations I've seen during that time.

The point of note is, percentages depend upon the basis they indirectly reference, you can't perform operations on them out of order and expect to get a accurate answer.

> The annualized return of IEF ETF ...

The annualized returns of bond ETFs cannot be relied upon because they all utilize shennanigans such as the yield to maturity loophole, and they do not mark to market the actual value of assets held in the ETF even when they expire.

There is also synthetic share manipulations on the options chains for those at times where the valuation and share price become entirely divorced from reality, also you don't get coupon payments on ETFs.

This is why, if you happened to be paying attention to the interest trends and purchased long-term inversely correlated contracts on a bond ETF like say... TLT ... well ahead of the FED announcing their intent to raise interest rates dramatically, you would have been significantly burned when the price didn't drop appropriately for the assets held in trust when they did actually raise interest rates.

Take a look for yourself, the TLT fund was ~99% 1.8 10YR treasuries trading at 120, FED raised rates in 2022. The value loss on the assets of the ETF put the ETF book value on a par ~$67/share in 2022 instead it remained between 120-110.

There are a lot of older people whose financial advisor told them to invest in bonds, and that market was teetering on the edge ever since 2020, and is now being backstopped through currency devaluation through blackrock/Fed partnership. This only creates more inflation, on top of the petrodollar/geopolitic adverse consequences.

Re: See how a dollar would have grown over the past 94 years [pdf]

#134
post #130

Earlier quoted context omitted.

> The chart you linked isn't inflation indexed. Neither was your comment. The alternative reading "You lose money from the inflation when the rate of interest is below the inflation rate which it almost surely was given the several decades of almost zero low-interest REAL rates in that time period" wouldn't have made much sense. The interest rate is below the inflation rate when the real rates are negative - not just…

> Neither was your comment. The comment I made refers to the underlying process of inflation indexing a yield curve in loose terms. You seem to be nitpicking on semantics for no real reason. You also seem to have messed up your math with regards to calculating the loss in purchasing power due to inflation of the principal and interest (coupon) of an investment. Inflation in 1990 was 5.4%, the return was 8.2% for that…

> The annualized returns of bond ETFs cannot be relied upon because

The annualized returns are calculated from the prices you pay when you buy and the price you get when you sell. Whatever you think they do seems irrelevant. The fact is that you put $100 in 2002 and you get $250 in 2020 (if you reinvest the money they give you every month - if you spend it you will have only $140 left but that doesn't seem the right way to calculate things).

> also you don't get coupon payments on ETFs.

You get distributions. You know that, right?

> Take a look for yourself, the TLT fund was ~99% 1.8 10YR treasuries

That would be remarkable for an ETF that "seeks to track the investment results of an index composed of U.S. Treasury bonds with remaining maturities greater than twenty years." Where do you suggest that I look?

The price of the TLT share has dropped more than 50% from July 2020 to October 2023. Imagine if they had not been manipulating the price! (In total returns terms it's slightly better but not that much: a 48% drawdown.)

Re: See how a dollar would have grown over the past 94 years [pdf]

#135
post #130

Earlier quoted context omitted.

> The chart you linked isn't inflation indexed. Neither was your comment. The alternative reading "You lose money from the inflation when the rate of interest is below the inflation rate which it almost surely was given the several decades of almost zero low-interest REAL rates in that time period" wouldn't have made much sense. The interest rate is below the inflation rate when the real rates are negative - not just…

> Neither was your comment. The comment I made refers to the underlying process of inflation indexing a yield curve in loose terms. You seem to be nitpicking on semantics for no real reason. You also seem to have messed up your math with regards to calculating the loss in purchasing power due to inflation of the principal and interest (coupon) of an investment. Inflation in 1990 was 5.4%, the return was 8.2% for that…

> You seem to be nitpicking on semantics for no real reason.

I guess I didn't interpret your statement about "several decades of almost zero low-interest rates" between 1990 and 2020 loosely enough.

> You also seem to have messed up your math

Your only issue seems to be with taxes and I explicitly mentioned that I didn't consider them, just like I didn't consider reinvestment until the end of the ten years period and I used rounded down yields and rounded up inflation numbers.

You say taxes can be anywhere from 10% to 37% - but choose to apply 37% to show the "error".

Let's consider taxes. 40% if you want, why not.

You buy in January 1990 a 10-year bond with a 8.36% coupon trading at par for $1k [for simplicity I use the monthly averages from https://home.treasury.gov/system/files/226/tnc_qh_pars_1.xls]. You get $83.6 per year, after taxes at 40% you keep $50.2. Let's say you put them in a box because you are too lazy to reinvest them. You will get then the $1000 back for a total of $1502.

$1000 in January 1990 are $1325 in January 2000.

$1502 are more than $1325. (And that's with high taxes and without reinvestment.)

--

For 2000-2010 (6.88%) you have $1000 + 10 $68.8 (100%-40%) = $1413.

$1000 in January 2000 are $1284 in January 2010.

$1413 are more than $1284. (And that's with high taxes and without reinvestment.)

--

For 2010-2020 (3.87%) you have $1000 + 10 $38.7 (100%-40%) = $1232.

$1000 in January 2010 are $1190 in January 2020.

$1232 are more than $1190. (And that's with high taxes and without reinvestment.)

--

It seems that you agree that over the three decades being discussed bonds did better than inflation (even after taxes) so I'm not sure why do you think I messed something up or what were you trying to explain to me.

> You still haven't broken even in purchasing power from the previous year

You're ignoring that the bond has appreciated from the previous year keeping purchasing power intact.

Re: See how a dollar would have grown over the past 94 years [pdf]

#136

Earlier quoted context omitted.

This is almost the definition of technical analysis: “chart always reverted to mean so it will always revert to mean” Note that almost every exchange outside the US has been flat or negative for decades. The US has held a precious position for a few generations that’s made “chart go up” feel like a given

> Note that almost every exchange outside the US has been flat or negative for decades. And the US itself was flat for over a decade, with the only thing saving a domestic investor's returns being bonds: * https://www.forbes.com/sites/investor/2010/12/17/the-lost-de... And as a Canadian, there are different sectors that would have given me positive gains over the years (I generally own mostly VEQT, a globally-diversi…

The thing is, we do actually have a tech industry and other countries largely do not. It’s like arguing that industrialized nations are only wealthier than agricultural nations if you include manufacturing. Thats the whole point!

Re: See how a dollar would have grown over the past 94 years [pdf]

#137

Earlier quoted context omitted.

I'm not sure that a US government bond has a meaningfully different risk profile than an aggregate investment in US equity markets.

If this were really true, they would offer similar returns.

Famously, Mr. Market deals in what things cost, not what they are worth.

Re: See how a dollar would have grown over the past 94 years [pdf]

#138

Note that the government has huge incentives to downplay the inflation rate. Over the time period of the graph, the price of gold went up by about 100x; I would consider that a more accurate estimate of inflation than the 18x number implied by the chart. Due to the way exponentials work, you can hide a lot of inflation over 100 years by claiming the annual rate is just half a percent lower than it really is.

> Note that the government has huge incentives to downplay the inflation rate. Over the time period of the graph, the price of gold went up by about 100x; I would consider that a more accurate estimate of inflation than the 18x number implied by the chart. The CPI is one of the most studied and examined statistics that the government releases. Everyone from hyper-capitalist financial traders to leftist unions folks (…

Your own link, to the BPP Wiki page, says:

> In 2007, the BPP initiated Inflación Verdadera, a project that provided a daily inflation gauge for Argentina, serving as an alternative to the official consumer price index which was manipulated by the government during 2007-2016

Re: See how a dollar would have grown over the past 94 years [pdf]

#139
post #38

Earlier quoted context omitted.

> This isn't useful or correct The formulas do not help you at all with the knowing. or not knowing, with being able to "predict" the past vs. being able to predict the future! They make assumptions. I would make the claim my statement is useful for what I said, which was for somebody looking back at a decision of the long ago past with hindsight knowledge. The post was not about somebody evaluating different investm…

The vast majority of your argument's volume was promoting an incorrect way of thinking comparing two unrelated things, for a question not asked or really that useful. You can't determine logical truth in a stochastic environment except after the fact when there is objective measure; and importantly there is no personal harm in doing this either, which is a direct contradiction to what you said. You then went way out…

> The vast majority of your argument's volume was promoting an incorrect way of thinking comparing two unrelated things

Funny, that is exactly what I saw when I read your "arguments".

> Any reasonable reader would throw away the vast majority of what you had to say as useless

You are talking about your own baseless attacks. You don't even try to argue, you replace coming up with arguments completely with going on the attack with otherwise content-free words, just accusations and insults.

Post reply on HN