Living Off Your Money by Michael McClung

Ask your investment, budget, and other money related questions here
Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Living Off Your Money by Michael McClung

Post by Scott 2 »

I picked this up off several recommendations. I am going to work my way through. I thought I'd start a thread for discussion and accountability.

First impression - oh crap, this is a textbook. The math doesn't look overly daunting, but it is going to be a slog. No wonder I couldn't find an audiobook.


2Birds1Stone
Posts: 1596
Joined: Thu Nov 19, 2015 11:20 am
Location: Earth

Re: Living Off Your Money by Michael McClung

Post by 2Birds1Stone »

I'll join you for the reread. Have to locate my physical copy.

nomadscientist
Posts: 401
Joined: Fri Mar 13, 2020 12:54 am

Re: Living Off Your Money by Michael McClung

Post by nomadscientist »

Looks rigorous. Although much of this analysis can be reproduced for free, I feel like he deserves my money.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

I started reading this weekend and made it through the first two chapters. Impressions:


Chapter 1 - I started learning about investing in the early 2000's. I bought into index investing strongly, down to the target retirement fund model. For about a decade, I've had the sense it isn't a great approach, but knew my high savings rate was outstripping any returns. So I didn't worry about it.

McClung nails me in a page - covering "The Revolution in Financial Economics". I guess the basic indexing strategy was outdated before I learned it. At the time, my sources were pop books like Rich Dad Poor Dad, The Millionaire Next Door, etc. It's not surprising I picked up a mediocre framework.

His Modern Mechanics section offers a really concise representation of ideas I've been glimpsing in more modern sources. Posts here, Tyler's portfolio charts, ERN, etc. I'd tried to bring them together organically, reading a variety of investing books, but never to connect all the dots. He does it in a page, obliterating my old mental model.


Chapter 2 - Due to my profession, I have a strong understanding of risk analysis, data mining, bias, etc. My experience is that a good framework is 90% of successfully working a problem. This chapter is fantastic.

The risk framework he provides, and summarizes in a single page, is far richer than my old indexing model. It is easy to understand and gets at something I've really struggled with from online sources. I tend to pick up small ideas - a blog article here, a forum post there, but find myself overwhelmed with the possible permutations. He puts it all together, providing a structure I can hang these concepts from.

One of my worries had been - "What if the US is now a declining empire? How can I trust back testing against US data?" McClung addresses it. He aggregates multiple global data sets, including Japan! He acknowledges the important and difficulty of building up these sources.

Equally important, he recognizes modeling bias AND explains it well. I was so happy to read this sentence:
The way to reduce the risk is by verifying results using an independent dataset (i.e., out-of-sample data) without tuning the strategy
For anyone in the field, this is an obvious concept. In my experience, however, it is not intuitive. It's a great sign he's using a strategy I both understand and believe in.


So far - I already feel the book has paid for itself. I wish I'd started here, instead of powering through more blog posts and forum pages. The thread on bogleheads really fixates on his prime harvesting concept. The book has so much more to offer. Hopefully subsequent chapters deliver on my now high expectations.

I also found myself collecting other authors for future reading. Jim Otar and Wade D. Pfau stood out on the bogleheads thread. I am prone to fixating on a topic, but wasting my time, by reading a bunch of books that say basically the same thing. Recognizing this weakness, I've committed to fully consuming McClung's book before pursuing other sources.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 3 - McClung reviews income harvesting strategies and introduces prime harvesting. He back tests for minimum safe withdrawal rate over 4 different real data sets, as well as generated data. He looks at success rates below 100%, off handedly taking 90% as a reasonable position for setting SWR. I always pick 100%, hmmm.

What stood out to me - a lot of really good ideas break down in the testing, especially when non-US data sources are introduced. It has me far more cautious about tweaks to tested strategies, that are not also subject to the same rigor.

He also casually disregards a rising equity glidepath, essentially saying "of course it performs better, you have more stocks". Given our current high CAPE environment, I'd found the rising equity model very appealing, but I also know CAPE is not the end all of valuation. McClung determines the data for CAPE is not sufficient, to effectively test for valuation in income harvesting rules. As a result, I feel less inclined to blindly adopt a rising equity glide path.


Re-reading the ERN article, I note two prime harvesting objections:

1. Nature does not make jumps. The McClung equity sell events are large, infrequent and occur at hard dollar amounts. ERN compares final portfolio value to WR. He finds stepping instead of a smooth line. He also proposes an improvement, but it is not subject to the same extensive testing as McClung's original model. Given how McClung shows other smart models break down in testing, I would need to see ERN's McClung smooth subject to the same checks, before adopting it.

Not to say that it isn't better, just that a few cherry picked years and a graph against one US data set isn't enough validation. Given my relative ignorance of the problem space, blindly trusting the change does not make sense. The promised benefits are not sufficient for me to pursue learning to test.


2. Prime harvesting violates the principle of optimality. Since the strategy has a memory of starting values, two retirees with the same age and holdings could have different withdrawal patterns. This is raised in the comments, drawn from the bogleheads thread. ERN acknowledges the limitation, but downplays the importance. I tend to agree. It feels like a theoretical flaw, opposed to practical concern.


One of my big considerations with any strategy - how can my wife implement this? Prime harvesting has a big win here, in that McClung distills it to a spreadsheet. If not her, I'm sure I could clearly document the process for a helper to follow. It does seem seem substantially better than blind rebalancing, especially in data sets beyond the US - ie the declining empire scenario.


One thing that bugs me, totally unrelated to McClung - Tyler put a ton of effort into the portfolio charts site, which is framed entirely around a static, rebalanced asset allocation. I have trouble reconciling that with the demonstrated benefits of a dynamic asset allocation, as show by McClung (and ERN). I have to believe Tyler is aware of the idea, so there must be some conceptual gap in my understanding. I have not read McClung's chapter on asset allocation yet, so that may end up offering a clue.


I think I have a better understanding of why the bogleheads thread centers on prime harvesting. It is the last idea in the last free chapter. Many commenters refused to buy the book. Humans are going to human, I suppose.

ertyu
Posts: 2893
Joined: Sun Nov 13, 2016 2:31 am

Re: Living Off Your Money by Michael McClung

Post by ertyu »

Scott 2 wrote:
Sun Feb 07, 2021 11:03 am
I have to believe Tyler is aware of the idea, so there must be some conceptual gap in my understanding.

Not really. Dynamic asset allocation does perform better and most who are not financially naive know this. Fixed asset allocations are for the benefit of those whose main priority - or skill - in life isn't financial or quantitative. It's to give normal people a simple heuristic they can understand, apply, trust, and stick to. Active asset management requires active study of the markets and the economy. It also introduces decision-making, and with that the whole host of buy high - sell low psychological errors that plague most inexperienced and unsophisticated traders. A fixed asset allocation, on the other hand, is something which at least in theory takes the decision making out of your hands: the decision is pre-made, your stock bond allocation will be 75/25; your task is to notice when it isn't, fix that, and "believe"

Fixed asset allocations thus might perform better not because they test better; they might perform better for a certain type of investor - which is most people.

ertyu
Posts: 2893
Joined: Sun Nov 13, 2016 2:31 am

Re: Living Off Your Money by Michael McClung

Post by ertyu »

Scott 2 wrote:
Sun Feb 07, 2021 11:03 am
I have to believe Tyler is aware of the idea, so there must be some conceptual gap in my understanding.

Not really. Dynamic asset allocation does perform better and most who are not financially naive know this. Fixed asset allocations are for the benefit of those whose main priority - or skill - in life isn't financial or quantitative. It's to give normal people a simple heuristic they can understand, apply, trust, and stick to. Active asset management requires active study of the markets and the economy. It also introduces decision-making, and with that the whole host of buy high - sell low psychological errors that plague most inexperienced and unsophisticated traders. A fixed asset allocation, on the other hand, is something which at least in theory takes the decision making out of your hands: the decision is pre-made, your stock bond allocation will be 75/25; your task is to notice when it isn't, fix that, and "believe"

Fixed asset allocations thus might perform better not because they test better; they might perform better for a certain type of investor - which is most people.

ertyu
Posts: 2893
Joined: Sun Nov 13, 2016 2:31 am

Re: Living Off Your Money by Michael McClung

Post by ertyu »

Scott 2 wrote:
Sun Feb 07, 2021 11:03 am
I have to believe Tyler is aware of the idea, so there must be some conceptual gap in my understanding.

Not really. Dynamic asset allocation does perform better and most who are not financially naive know this. Fixed asset allocations are for the benefit of those whose main priority - or skill - in life isn't financial or quantitative. It's to give normal people a simple heuristic they can understand, apply, trust, and stick to. Active asset management requires active study of the markets and the economy. It also introduces decision-making, and with that the whole host of buy high - sell low psychological errors that plague most inexperienced and unsophisticated traders. A fixed asset allocation, on the other hand, is something which at least in theory takes the decision making out of your hands: the decision is pre-made, your stock bond allocation will be 75/25; your task is to notice when it isn't, fix that, and "believe"

Fixed asset allocations thus might perform better not because they test better; they might perform better for a certain type of investor - which is most people.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

A fair perspective. I know very few people IRL who I could convince to read and understand this book.


Chapter 4 - McClung surveys variable withdrawal strategies, introduces a comparison metric, then tests them heavily. The author's exclusion of direct valuation continues, meaning the CAPE based rules I found appealing are not discussed.

The chapter is math driven, but he provides a good qualitative discussion of each approach. It becomes clear that (for a given level of risk) naive fixed amount or fixed percentage withdrawals leave a lot of potential spend on the table. The latter is also subject to unacceptable volitility. Equally important - just because something seems like a good idea, that does not mean it tests well. He fails his fanciest model.

I appreciated the eventual firm recommendation of a single strategy, along with the provided spreadsheet for calculations. I doubt I'll find a better overall approach, given limitations of my personal expertise.


Math is boring, so I cheated and skimmed ahead. Bringing together the breadth of the book's strategies, he believes 55% more retirement income is achievable - relative to the simple indexing, rebalancing, minimum safe withdrawal approach.

That's a big promise. I am motivated to keep going. While I am not in need of the extra withdrawals, the added safety is appealing. The return on effort is high. It also appears accessible, with a low maintenance effort.

Funny - I recall disregarding Jacob's posts about this very idea. I remember thinking, sure it's easy for him, but he wrote the book on ERE. How much improvement can the rest of us expect to eke out? I read Bogle, after all...

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 5 - McClung sets to defining how we evaluate a retirement portfolio. Taken blindly on faith, this chapter is easy. He provides a harvesting ratio metric, it ranks portfolios from high to low. Done.

Trying to understand the why. I'm still not sure I do. He jumps right into the deep end, with mean variance optimization and the efficient frontier. This about the bounds of my investing education.

In my shattered world - I buy a portfolio on the efficient frontier, with volatility my time horizon will tolerate. I'm investing for a long time, and I like money, so that's easy. Stocks to the moon.

Only the strategy tests as mediocre during drawdown. F. The most desirable withdrawal patterns don't live on the efficient frontier.

He adds on the capital asset pricing model, market portfolio, and Sharpe Ratio. I've read about these, but never think of them in any intuitive way. My ignorance is ok, since his testing shows no useful correlation to good drawdowns.

Except, then he drops the bomb that the total market fund approximates the portfolio with the maximized Sharpe Ratio. Dammit, that's basically the portfolio I copied from early 2000's retirement books.


The rest of the chapter introduces and tests his metric. It correlates well with positive draw down experiences. He already broke my framework, so I am easily convinced. I certainly don't know enough to challenge or improve upon the harvesting ratio metric.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 6 sets up a baseline market and portfolio for testing retirement portfolio construction. I don't think it is intended as any sort of revelation, but I apparently am still clinging to the "simple path to wealth" paradigm. Noteworthy:

- Great conceptual explanation of why a global portfolio is desirable. The "US companies operate internationally" line is weak in comparison. Especially after he reinforces it with testing.

- A reminder that testing includes low return worlds. I think this is us today. My instinct is to calculate a MSWR and plan to spend a percent under that. He knows.

- He casually throws out a table of MSWR for various equity portfolios. These are not combined with bonds, but the US total market underperforms, even relative to his naive baseline portfolio.

His baseline portfolio reduces the large cap exposure vs. US total market, so I suppose this isn't a surprise. But that's exactly the point, highlighting the flaw of naive indexing. I'm a little nervous about historical small value premiums distorting results, but I read that somewhere and don't have a better idea.


I note the author is ignoring bond allocations almost entirely for now. I can see that is one possible way of combing the tools on Tyler's site with the prime harvesting strategy (dynamic asset allocation). I imagine a better model will be offered as the book progresses.


While this learning is important, it doesn't feel particularly good. My investment paradigm is reshaped by every chapter. I don't enjoy being so wrong or facing my ignorance. Probably a lesson there.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 7 covers setting the initial stock percentage of the retirement portfolio. I've struggled with the traditional recommendations for early retirees - upwards of 80 to 90%. That's the only way the money will survive so long, after all. Scary with CAPE over 30. A rising equities glide path gets at this fear, but I still wonder about risk tolerance as I age.

McClung shows a dynamic asset allocation strategy changes the picture. His testing suggests prime harvesting works with much lower stock percentages. It is also less sensitive to variations in the starting stock percentage. He recommends 55-65% for a retiree with a 40+ year term. I can sleep better at night with that, for sure. Not to say prime harvesting couldn't also produce an equities heavy portfolio, but knowing how I got there might make it easier to accept.

He briefly touches on valuation in setting the stock percentage, essentially saying it won't matter much, since income harvesting will balance things out. Proof is promised in Chapter 9.


I'm a little troubled by his general comfort with high bond allocations, given with current interest rates, they aren't going to appreciate. I'm guessing he touches on this more later. He's also ignored gold so far, other than generally dismissing commodities, for following different rules. I already have some personal reservations regarding gold, so this isn't a huge problem for me, but I do wonder about it.

ertyu
Posts: 2893
Joined: Sun Nov 13, 2016 2:31 am

Re: Living Off Your Money by Michael McClung

Post by ertyu »

following along as you type these out. dropping in to say i share the same reservations: if (1) high cape predicts low future returns on equities, and (2) sovereign bubble, bottom of LT interest rate cycle, inflation/YCC to dead with debts, etc., what does this mean for a traditional 60/40?

all strategies discussed seem to be a variation on the 60/40 idea - they seem to, in essence, be techniques for fiddling with the stock/bond allocation across time. but what if there's good reason to believe that both debt and equity have poor prospects?

Smashter
Posts: 541
Joined: Sat Nov 12, 2016 8:05 am
Location: Midwest USA

Re: Living Off Your Money by Michael McClung

Post by Smashter »

I am also following along and appreciating the detailed reviews. Thanks for doing this! I look forward to hearing his take on high bond allocations.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Good to know this is benefit someone in addition to me. @ertyu - I have no idea how McClung would answer your question.

Chapter 9 covers portfolio construction.

I enter into this chapter ignorant, at best. My investment holdings are life strategy funds (diversified over several, lol). I also have a moderate growth fund from my 401k. Bogle convinced me to blindly buy his product. Fortunately, a lot of other people have too, so it's been ok.

I expected McClung to continue down the randomly generated portfolio path, but he does not. Rather, he uses harvesting ratio to iterate on lazy portfolios from Simba's spreadsheet:

https://www.bogleheads.org/wiki/Simba%2 ... preadsheet

Tweaks are small. Adjusting to similar asset classes, so he can get a longer (40 year) set of back testing data. Normalizing bond allocations, so stock weighting doesn't skew relative performance. There's a bunch analysis. I note:

- His tests favor owning all the things, equally weighted, rather than market weighted (ie 10x10 portfolio style). He notes lazy portfolios are not designed for drawdown, even if retirees use them this way.

- Undiversified portfolios (ie 50/50 small value and emerging markets) test well under his criteria. He does not recommend them, due to lack of diversity. I've read about a premium on small value stocks that no longer exists, but could skew back testing. This might be a good example of accounting for it.

- The portfolios that test well avoid heavy allocations to large stocks, especially large growth. With my buy the market indexing strategy, this is what I own. Yay market weighted investments.

- After all that effort developing harvesting ratio, he treats it as a rough guide, not an absolute measure.

- He acknowledges US market risk (my portfolio again). He proposes addressing it, by limiting US stock allocations to 30% of equities. Then he puts the extra into REIT's and emerging markets. I barely know what those are (houses! India!), so it's hard to have a reasoned response.

- He highlights that owning a "large value" index fund, doesn't mean only owning large value stocks. He calculates the true percentage of each asset class category, for each recommended portfolio. This requires digging into what the underlying funds own. He provides a chart, comparing these allocations to baseline market portfolios.

The charts highlight how undiversified my "own everything" diversification strategy is. US large stocks are dominant. His favored portfolio stands out as weighted on emerging markets and REIT's. He subdivides the two asset classes later, to address this concern.

- Bonds are hand waved away again - "there is little to add to what others have already said well in other books." He gives a simple recommendation - 50% TIPS, 25% short term treasury, 25% intermediate term treasury. Seems legit? I have no idea, or even what the other books are. For the amounts I plan to own, a little more education seems prudent.

- Commodities get a footnote - he can't test them, but up to 10% of equities allocated to commodities (ie gold) would be reasonable.


I had to read and re-read this chapter. I went to portfolio charts and tried some of the recommended allocations. They seem to test ok, but no better than the canned portfolios. Adding gold makes them "better" using Tyler's metrics. Nothing looks as good as his golden butterfly portfolio. However, his data and metrics seem to favor a small value tilt and gold bias. Not saying that is right or wrong, just observing it.

Now that I've read more, Tyler's site doesn't look like an expert blessing the masses with his wisdom. I doubt he ever intended that, but in my relative ignorance, that is how it felt. I now see it as someone sharing his learning with the community.

Removing the pedestal, it makes sense that he hasn't covered dynamic withdrawals. That's a hard thing to model, especially given what he offers. The charts are also covering the full breadth of FIRE (ie accumulation and withdrawal). One size fits all portfolios, for those disparate needs, now seem impractical. Taking money out is hard on a portfolio.

I also tried running one of the portfolios in Simba's spreadsheet, to understand it. The metrics are similar to Tyler's site. Interesting. Further evidence he is sharing learning with the community.


The final portfolio recommendations own a lot of stuff. 13 distinct funds in each of the top two. When I start thinking about implementation over the breadth of my accounts - ugh. Rebalancing each year seems like an enormous pain. I am not committed. I may seek mutual funds that approximate the stock and bond allocations. I could then limit my attention to the variable withdrawal and income harvesting strategies.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 9 focuses on setting initial withdrawal rate at retirement. McClung finally accounts for valuations.

He gives evidence a single metric isn't reliable enough, then provides a mechanism to aggregate several US metrics - CAPE10, Q Ratio, iCAGR20. He later adds CAPE10 for emerging and developed markets to the mix.

After all that, he only commits to rough banding of the initial withdrawal rate. Possible values range from 1 to 4. High valuations start at a 3.8% rate, with low valuations going as high as 5.5% for his baseline case. The spreadsheet on his website values today's market at 2.9, brought down by iCAGR20 and ex-US valuations.

His baseline withdrawal rates are for a 30 year term. For a longer retirement, the range is more like 3.4% to 4.6%.


My main takeaways

- valuation is a very rough metric

- international diversification is encouraged with today's valuations

- His numbers put me around a 3.9% initial withdrawal rate, ignoring any impact of social security. That feels too good to be true, but I follow the logic. In comparison, the CAPE based rule was giving me 3.22%.

- For the recommended strategies, he views the calculated initial withdrawal rate as conservative, with several margins of safety built in.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

Chapter 10 is a rehashing of McClung's recommendations. I was paying attention earlier, so relatively few notes:

- Cap annual withdrawals at 1.5x inflation adjusted initial withdrawal amount, using his preferred variable withdrawal strategy

- Stock allocation for a longer retirement is set at 55%, opposed to 50%.

- He shys away from chasing more income with higher stock allocations, but does discuss the impact - higher volatility and less remaining money at the end. The latter is symptomatic of a higher initial withdrawal rate and his variable withdrawal calculation. He is especially against it when valuations are high.

- Leaving money behind is a bequest. He talks about how to do that. Not my priority, but the rules are common sense.

- He shows one time unexpected events have limited impact on success rates. The variable withdrawal strategy absorbs the impact well.

- He's not opposed to restarting the withdrawal calculations partly through retirement. I could see that happening in event of an inheritance or unexpected work.

- He generally believes the aggregate strategy is robust to small tweaks, which matches how he was building the framework.


Overall, I'm going to adopt a lot of what he's offered. My biggest hesitations are related to asset allocation - REIT's, emerging markets, and bonds in a zero interest rate environment.

I also hesitate to spend at 3.9%+ withdrawal rate. Easy to argue, after accounting for my social security, that number should be 4.15%. It feels very counter intuitive, even though I understand his testing.

I'm going to finish my other couple books, then track implementation in my journal. I don't know how to resolve those reservations yet.



Chapter 11 - The last 80 pages of the book center on guaranteed income. He doesn't recommend considering it until 60. Even at that age, he only seems to like CPI adjusted immediate annuities, which you can no longer buy.

The value proposition changes for someone as they get older, when it makes more sense to pool mortality risk with others. This matches the bogleheads VPW strategy, which also uses annuities as the end game.

Other than recognizing how valuable the inflation adjusted nature of social security is, I'm too young to worry about guaranteed income. More so, since interest rates are rock bottom, which ruins his strategies, except at very old age.


The book was easily worth my time and money.

ertyu
Posts: 2893
Joined: Sun Nov 13, 2016 2:31 am

Re: Living Off Your Money by Michael McClung

Post by ertyu »

Curious to see what ultimate strategy you arrive at for yourself. Thank you for these write-ups.

Scott 2
Posts: 2824
Joined: Sun Feb 12, 2012 10:34 pm

Re: Living Off Your Money by Michael McClung

Post by Scott 2 »

I found some limitations with content in the book/spreadsheet:

1. The spreadsheet only covers a 50 year retirement. Not a big deal to extend it to 60, but I was getting an error until I figured that out.

2. Several of the funds used for the sample portfolios have issues:

2.1 Vanguard stopped offering investor share classes for many funds. McClung is using the investor tickers. For the most part, it's not too hard to map these to the admiral share classes.

2.2 A couple of the small and emerging markets funds are now liquidated. This highlights a risk / limitation of slicing the portfolio as he suggests, that I didn't fully appreciate.

2.3 Maybe not an issue, but most of the recommended emerging market and international funds are actively managed. With a little consideration, the reasons seem obvious, but it took me by surprise. Investing in these categories means you are selecting a management team.


Overall - I would say McClung is largely (and probably intentionally) opting out of the mechanics around assembling a portfolio. Coming from a "buy the life strategy fund" perspective, I am woefully unequipped here. More learning to do.

Post Reply