The 4 Percent Rule is Not Safe in a Low-Yield World

[Egg]

A lot of international guys are blindly adopting the US numbers. Worse, it seems that very many think that 30 years = infinite number of years.

Even there, I don't think there's necessarily a great cause for worry. I am UK-based, and initially made exactly that mistake of thinking this was a purely mathematical challenge to get to the golden 4%. However, my consious interest in ERE began a couple of years ago, and in the intervening time my thinking has changed somewhat. Even for "hardcore" ERE (≤5 years) there's enough time for a fair amount of introspection and reevaluation of goals.

[steveo73]

Even for "hardcore" ERE (≤5 years) there's enough time for a fair amount of introspection and reevaluation of goals.

I think that this is probably the least safe way to retire. You do have to cut down your expenses but if you are too tight and somehow come up with a proposed WR that looks good on paper however it doesn't last into the future than you could be in trouble.

I'm mostly concerned about late life medical and other care because who knows what will happen.

[IlliniDave]

I'll be ~mid-fifties when I retire, so 30 years is likely to be plenty of time from an actuarial perspective. Nevertheless, I'm planning a two-tiered approach where my current level of spending falls at or below 4% (it represents having "more than enough"), and my threshold level (as frugal as I can imagine without really hurting) falls below 3%. My actual plan calls for something in the middle and even then I hope to be able to exceed my plan (exceed plan = spend below). The biggest thing IMO is to leave yourself some flexibility and avoid planning for just barely 25X of you drop-dead, no-kidding, minimum financial threshold. If you can comfortably adjust downward for a time to ride out a few rotten years you should be able to carry on.

I agree with steveo73 in the sense that my bigger worries are what happens in the out years when I will be less capable of of doing things for myself (and the possibility of going out and finding employment is near nil) and more reliant on money for things like healthcare and home/property maintenance and the like. My intent is that with my "nominal" plan that my spending from assets is low enough to allow growth of my assets and wind up with more wealth at 80 than say 55. I might not achieve that, but a prerequisite for me to disconnect is being able to judge there is a reasonable likelihood of that happening.

[jennypenny]

During the last few years, other expenses would probably drop, giving you a little wiggle room. Things like travel, outings, hobbies, and the need for a car all but disappear. Even grocery expenses drop as caloric needs diminish. That will help offset some of the increased medical costs.

I'm most concerned about housing costs during that time if we can't manage living in our own home anymore. We need to talk Ego into opening a Soylent Towers for EREs.

[GandK]

I'm most concerned about housing costs during that time if we can't manage living in our own home anymore. We need to talk Ego into opening a Soylent Towers for EREs.

I had a good giggle this morning imagining all the regular forum posters in the same old folks home.

[penny]

What is a safe amount to take from a pension (UK):

http://www.schroders.com/en/uk/private- ... a-pension/

[BRUTE]

I had a good giggle this morning imagining all the regular forum posters in the same old folks home.

get off brute's lawn! *quivering voice, cane shaking*

[tommytebco]

Any discussion about "safe" is always bordered by several "unthinkable" states. These are Armageddon, total financial collapse and?? maybe the last coming.
Anyway, the way I see it, set a plan, start it, review annually, (or as required). If you feel like a "safety engineer" (they wear a belt and suspenders because 'you never know?' you will save up a much bigger pile, thus delaying your retirement, but in the event of the BIG SHTF it makes no big difference.

For most real life situations, the worst down side to too early is sitting in your seat saying "Welcome to WalMart" from time to time.
Not the end of the world.

[Jean]

I feel safe at 20%

[jacob]

The latest newsletter from AAII points out that FOMC's current long term(*) growth forecast of US economic growth is now down to 1.80%/yr (real return). The forecast has been declining every single year for the past five years (it was 2.5% in 2011). Add to that the current 2.05% yield of the S&P500 as a representative market yield, at we get expected total stock market returns of 3.85%/year. Of course returns could be higher if valuations increase even further or much lower if valuations revert to historical values.

(*) An ill-defined timespan as far as I can tell, but more than 3 and typically 10+ years. (The models work off of the 5 and 10yr treasury market rates.)

TL;DR - The 4% rule is currently expected to be threading water/sinking very slowly over the next decade.

[ThisDinosaur]

How are these forecasts made? Have their methods been consistently reliable in the past?

[7Wannabe5]

Doubt they included my potato harvest in that model.

[jacob]

@ThisDinosaur - As far as I can tell, each [of the 12] member has their own econometric model that's likely based on stuff like inflation rate, unemployment numbers, market rates, ... Then the FOMC combines those forecasts which is how you get the 1.85%. Over the past 5 years, the forecast has been overly optimistic compared to reality. I suspect due to the way that the input is too rose-colored compared to reality (e.g. inflation is higher than reported, unemployment is higher than reported, ...) for various political or cultural reasons.

[ThisDinosaur]

So its an average of 12 expert guestimates? If we knew their individual models, and could substitute more reliable numbers for the variables (inflation, unemployment) could this be used as some sort of arbitrage opportunity?

[A Life of FI]

Without getting into the correctness of the Fed's projections, I would add the cash spent on buy backs to the 3.85% (1.8% GDP growth + 2.05% dividend) total return number above. The buy backs are another form of dividend, which reduce the total number of a company's shares outstanding and thus increase the value of the remaining shares. Since companies are currently spending more money on buy-backs than dividends, this would add 2%+ to the total return number bringing up from 3.85% to 6% to 7%.

https://ddcvqwvagfyuw.cloudfront.net/co ... rnings.jpg

[ThisDinosaur]

I don't see that buybacks increase individual share value or total return. If a company buys its own shares, it owns a larger percentage of the incoming profits. Outstanding Shares + Company Owned Shares still add up to 100%, yes?

[A Life of FI]

@TD outstanding shares refered to shares which have been issued and not bought back by the company. Legally speaking a company can't own an interest in itself. When a company buys its own shares back they take on the same status as shares that were never issued, they have no rights to profits, no dividends can be paid on them, they can not vote in shareholder meetings, they are not entitled to any assets in the case of a liquidation of the company etc. Effectively for ownership purposes they do not exist.

Here's a link:
https://en.wikipedia.org/wiki/Shares_outstanding

[chicago81]

I don't see that buybacks increase individual share value or total return. If a company buys its own shares, it owns a larger percentage of the incoming profits. Outstanding Shares + Company Owned Shares still add up to 100%, yes?

When the company buys back its own shares, it essentially "retires" that number of shares from the float. The "value" of any shares that are "company owned" can be thought of as being distributed among all the holders of the Outstanding Shares.

It is meaningless to think of "Company Owned Shares" -- because all of the individual shareholders are the ones who OWN the company.

[jacob]

@ThisDinosaur - Indeed!

What we're looking at here is earnings growth + money plowed back into companies. Historically (20th century), US corporate earnings growth has been about 6% annually nominal over a business cycle. If you sum it all up and subtract inflation, you get a pretty good match to GDP growth real.

The dividend is something that leaves the corporation and goes to all shareholders. If all the shareholders invest that money in new companies or existing ones that money will make the economy bigger.

Conversely, a buyback is a company buying shares from some shareholder who now no longer has those shares; instead he has money. Now of course that shareholder could go and spend those money on buying the share backs again ... but if he does that the net transaction is a big fat zero. Essentially, buybacks is just money switching hands from corporation to shareholder. Such a switch does nothing to physically grow the economy. It's mere financial engineering which only serves to increase the stock price. The reason companies do it is that they get to reduce the number of outstanding shares. This makes EPS go up and therefore it looks like the company has become more profitable on a per share count even if the company didn't actually grow physically as they added a "financial department to the company which engages in the business of trading their own stock".

PS: Often this is done with borrowed money and so it isn't a sustainable solution (see e.g. IBM a couple of years back). You can see how buybacks tend to peak right before the stock market (or company price) crashes. Companies are remarkably good at buying high and selling low. It's extremely tempting to engage in these strategies because they leverage P/E which in turn goes even higher. The perfect definition of a bubble.

[ThisDinosaur]

I get that EPS goes up following a buyback. This should fool non-shareholders into buying the stock, increasing demand, and pushing the price up higher. The company now uses the new capital from dumb investors to grow the company, pay debt, give executive bonuses, whatever. What I don't get is how you can ADD the buyback cash to the total return.

Unless you can reliably depend on the price appreciating upward to an EMH equilibrium point. Is that the case? Do stock prices go up by a *predictable* quantity following a buyback?