On sloppiness v accountability

I was reading an article on golf in the weekend version of the WSJ and ran across this quote:

"I don't see the point of hitting a golf shot in practice without being accountable, given that every shot in competition, you're accountable in a round...as a behavior, it doesn't make sense to me."

This made me think of two things.  First, I was thinking that I generally feel like I am often in practice mode here at RH. I cut corners, I estimate, I get sloppy, I hack, I fudge, I elide, I leap to conclusions. etc etc. 

Then, the other thought I have is that it doesn't matter because the one thing that I am here is "accountable." For all the sloppiness and elisions there is no point at which I am not 100% accountable to my family and my future self.  The stakes and consequences for understanding what I try to figure out here could not be higher.   The work I do here may not matter to the world at large but it matters one hell of a lot to a few real people here in south Florida.

Some thoughts on modeling complexity

First of all, I'm making this up.  Second, this seems to be true enough to at least me after three years of doing retirement modeling and simulation. Let me try to dilate on the letter-points along my fake curve.

Testing last year's naive intuition on consumption utility with a lifetime utility model

Last year in this post (My Own Private Idaho Of Utility - A Case Study In Spending Control) --before I had figured out the idea of multi-period lifetime consumption utility modeling -- I had made a naive stab at creating my own single period consumption utility formula just for fun.  It was ugly and un-informed and more or less tongue-in-cheek. It looked like this: U(Ct) = 7e-11*x^3 -6.85e-06*x^2+.1396x-565.56 ...for reasons I can't recall but could if I re-read my own post.  The basic idea was that I wanted a hump-shaped function because I figured that spending 35k has utility and spending 36k has slightly higher utility and spending 37k has even higher but diminishing utility, BUT spending 100k might ruin you and might be wasteful to boot. The implication for me at the time was that consumption utility (if perpetuated over time) would not really rise as more is spent, it would go down at some critical point.  The typical formulas for consumption utility, however, like CRRA or log utility have monotonically rising and diminishing utility behavior so I made up a new formula (above) to get the hump.   I wanted:

A Weak Defense of the Constant Spend Assumption in Retirement Modeling

THE CHART

Since I have often ripped on constant spending as an unnecessary and foolishly high risk idea[1], let me try to explain my chart to set up the post...

Book Mention -- Willful Ignorance

Just finished Willful Ignorance, the Mismeasure of Uncertainty by Herbert I Weisberg, cr 2014, 452 pages.  This is not a review, just a mention, but it was, in the end, gladly read. The book can be roughly characterized as going to school on the history, philosophy, meaning(s), lacunae, and implications of our modern conception of probability theory.  It was also written by a neighbor of one of my retirement-quant interlocutors, Francois Gadenne. From Amazon's description (I am not an affiliate, btw):

Through a series of colorful stories about great thinkers and the problems they chose to solve, the author traces the historical evolution of probability and explains how statistical methods have helped to propel scientific research. However, the past success of statistics has depended on vast, deliberate simplifications amounting to willful ignorance, and this very success now threatens future advances in medicine, the social sciences, and other fields.Limitations of existing methods result in frequent reversals of scientific findings and recommendations, to the consternation of both scientists and the lay public.

Willful Ignorance: The Mismeasure of Uncertainty exposes the fallacy of regarding probability as the full measure of our uncertainty. The book explains how statistical methodology,though enormously productive and influential over the past century,is approaching a crisis. The deep and troubling divide between qualitative and quantitative modes of research, and between research and practice, are reflections of this underlying problem.The author outlines a path toward the re-engineering of data analysis to help close these gaps and accelerate scientific discovery. 

Or maybe it is pithier and more efficient to quote Mr Weisberg himself:

"...probability is not destiny"

But you'll have to read to find out why.

Alpha Architect on Investment Factors in Bonds

In this recent post, AlphaArchitect reports on efforts to deconstruct factor influence on corp bonds. To quote: "The presence of historical prices impacting future returns, i.e., momentum, has been well researched in the equity market, which we’ve covered here. We’ve also closely looked at momentum in bond markets here, here, and here. What the Bali, Subrahmanyam, & Wen are exploring is whether momentum shows up in the corporate bond market, and if so where?"

The conclusions are broader than what I put into this post but this point was interesting:

1. Some of the most interesting conclusions of this paper came in the robustness tests finding the sources of STR, MOM, and LTR.  When digging deeper into short-term reversals in the corporate bond market the researches found that STR disappears in the most liquid bonds and was strongest in the least liquid section of the bond market.  This indicates that there is a liquidity-based explanation for short-term reversals.  The researchers also found that the momentum effect was strongest in the highest default/credit risk bonds, while the factor becomes statistically insignificant in the lowest default/ credit risk section of the market.  In addition to those findings, they also found that momentum is strongest in times of economic downturns and periods of high default risk.  In fact, the impact of the financial crisis was so strong that when the financial crisis is excluded from the data the MOM factor becomes statistically insignificant.  Much the same as momentum the long-term reversal factor was strongest in the highest credit risk and highest default risk bonds.  In fact, the researchers found that credit rating downgrades are an important source of the long-term reversal effect.

While I don't have the research chops to critique, comment, or replicate, I find the feedback from the reported research to at least be confirmatory.  I can say that because:

1. The single biggest trade profit I ever received over an entire lifetime by way of an active time-delimited trading position made with existing capital was in high yield at the depths of the financial crisis in Q1 2009, something that was held for approximately 24 months thereafter.  While the at-the-time 25 point spread seems like a no-brainer in retrospect looking from our current perch in 2018, I will admit that at the time it took some courage to un-freeze from my death-crouch and commit meaningful and steadfast capital

2. It's been a while since I have done the analysis but since I trend-trade across credit risk categories I once took a look at what works best over what timeframe.  High yield etfs over 12-18 months won.  My take away was that they had equity-like behavior and high beta so I was not all that surprised. 

Top all-time RH posts as of July 17 2018

Summer doldrums so here are my top all time posts according to blogger.com


Ed Thorp and AQR on Retirement

The 2017 RiversHedge 20 Best-of Awards

On the affinities between hurricanes and retirement

Why, exactly, are people losing their minds over this correction? [1]

Saving Preferences After Retirement in US, Netherlands and Australia

Wealth Depletion Time - an Hypothesis and a Self-Challenge  [2]

Retire the Blog? not yet but...

Public Service Announcement from RH on fragile systems

Bitcoin madness

I am my own behavioral case study on annuitization

RH links - 7/1/18

QUOTE OF THE DAY

There are plenty of ways to get rich — start a business, save & invest wisely, inherit money, get lucky, etc. But staying rich involves just a few simple things — self-awareness, modesty, and the ability to delay gratification with a portion of your capital. Ben Carlson

RETIREMENT FINANCE AND PLANNING

How High Net Worth Individuals Invest: Their Asset Allocation Breakdown, Financial Samurai

it’s obvious you should start a business if you hope to get really rich one day. https://www.financialsamurai.com/how-high-net-worth-individuals-invest-asset-allocation-breakdown/

One More Advantage of Using the Actuarial Approach—No Sequence of Return Risk, Ken Steiner

the Actuarial Approach and Actuarial Budget Benchmark (ABB) advocated in this website is a dynamic approach that will avoid SORR.  It automatically recalculates the annual spending budget to maintain the balance between the market value of the retiree’s assets and the market value of the retirees’ spending liabilities.  http://howmuchcaniaffordtospendinretirement.blogspot.com/2018/06/one-more-advantage-of-using-actuarial.html

Some Risks Can't be Modeled, Dirk Cotton

The shortcoming of MC simulation is not that it will create unrealistic scenarios — quite the opposite — it won’t generate many highly unlikely outcomes. So, even after we test retirement plan risk with simulation we still don’t know much about the effects of low-probability catastrophic events…After simulations, we still need a way to plan for the unknowable…Avoiding unforeseeable risks is clearly not an option. It's hard to steer around an obstacle you don't know is there…The best spending rules won’t eliminate these risks. After a long sequence of poor returns, they will simply reduce safe spending to a level that no longer supports the household’s standard of living. Nor will the best simulation software ferret them out and suggest fixes….Low-probability catastrophic outcomes defy avoidance and mitigation but they’re worth contemplating and possibly worth insuring. http://www.theretirementcafe.com/2018/05/unpredictable-retirement-risk.html 

A Compendium of my "Wealth Depletion Time" Posts

I've been through a longish journey this year on the math and simulation of the lifecycle economics idea related to the "expected discounted utility of lifetime consumption" [in the presence of the potential for a period of time where wealth is depleted before the end of a (random) lifetime and where spending would then "snap" to available income].  The journey is not over since I will be successfully getting my software back from the Geeksquad after a 6 week wait for a recovered hard drive...and I have some ideas for next analysis.  In the mean time, let me summarize what I have so far.  I'm surprised by how the ending is not so different from what I hypothesized at the beginning. I'll attribute the whole effort to a reading of some work by Prof. Milevsky who emphasized both the importance and subtleties related to understanding the concept. I'm more or less glad I went down the path.  The effort has helped me "see" retirement finance in a way that I consider either comprehensive or integrative or both.  That was one of my goals for years.

The beginning

• Wealth Depletion Time - an Hypothesis and a Self-Challenge 3/24/18

The (current) end

• Wealth Depletion Time Simulator - an update and summary, 6/29/18

Interim posts

• Wealth Depletion Time - Adding again to my pile with Leung '94 3/31/18
• Wealth Depletion Time - Selected Excerpts 4/3/18
• Is this an easy "Wealth Depletion Time" analysis method by proxy? 4/3/18
• Wealth Depletion Time - Commentary 4/4/18
• Coming soon -- The Wealth Depletion Time Game? 4/4/18
• A trial run of a WDT game prototype 4/6/18
• Buying some extra "certainty equivalent constant" spending given a wealth depletion framework. 4/8/18
• Dividend investing meets Wealth Depletion Time and...nothing 4/9/18
• Optimizing Spend Rates in a simple WDT model 4/10/18
• Supplement to "optimizing spend rates in a WDT model" 4/11/18
• Sliver of hope for dividends?? 4/12/18
• 3D "utility surface" in a Wealth Depletion Time model: spend rates vs annuity 4/13/18
• 3D utility surface in WDT model: return vs spend rate 4/17/18
• Putting it all together: a wealth depletion time "utility simulator" 5/3/18
• The Wealth Depletion Time EDULC Simulator is live! 5/27/18
• Another way to look at the wealth depletion concept using two spend rules and charting real consumption, 5/29/18
• Snap-to-income scenarios can come from below not just above, 5/30/18
• What if advisors were paid based on RULC? A thought experiment 7/9/18

On Turning 60

I thought I'd share some blogger thoughts on hurdling age 60 this week. This is not deep philosophy just some random retirement-finance blogger comments.  The milestone has not only allowed me to shift my thinking from a posture of "early retirement" to, let's say, "more conventional retirement" but also to reflect on where I was then (before 60) and now am.

Epic father-daughter road trip, episode 2.

Here is episode 2 of my next installment of "epic road trip" with daughter #2.  I need to create a poll for votes on ideas for episode 3 in 2019 but all 5 of you readers can chime in as you wish in the comments. I'm thinking either north towards Alaska (not sure about the vehicle situation) or south towards California then east at SF or LA. I'm thinking SEA to LAX to Las Vegas to canyon country, and then back to DEN. An alternate is north then east across Canada but I've done that and Alberta+Saskatchewan, for the uninitiated, is about 400,000 miles wide.  And flat. If, in Winnipeg, one can see a runaway dog for 3 days, then up that to 10 days in mid-plains Canada where the surface of the earth doesn't even remotely curve. 

Dirk Cotton on Uncertainty

I was reading and excerpting a recent post by Dirk Cotton at The Retirement Cafe on managing uncertainty in retirement. My problem was starting to be that the excerpts were tending towards being, in effect, a copy of the whole post. So go read it...

Some Risks Can't be Modeled, Dirk Cotton

The shortcoming of MC simulation is not that it will create unrealistic scenarios — quite the opposite — it won’t generate many highly unlikely outcomes. So, even after we test retirement plan risk with simulation we still don’t know much about the effects of low-probability catastrophic events…After simulations, we still need a way to plan for the unknowable…Avoiding unforeseeable risks is clearly not an option. It's hard to steer around an obstacle you don't know is there…The best spending rules won’t eliminate these risks. After a long sequence of poor returns, they will simply reduce safe spending to a level that no longer supports the household’s standard of living. Nor will the best simulation software ferret them out and suggest fixes….Low-probability catastrophic outcomes defy avoidance and mitigation but they’re worth contemplating and possibly worth insuring. 

Uncertainty in Equity-Bond Correlation Assumptions

I was just reading this article on equity bond correlations (Challenging Equity-Bond Correlation Assumptions, winton.com). The gist of the piece is roughly this: 

what number should be used? The distribution is fairly flat and so any value between -60% and 60% is almost equally likely. By taking the average, we are simply picking a number close to the middle of the distribution rather than choosing the most frequently observed correlation...Often we assume future correlations will be similar to those experienced in the past. Naively, we may be tempted to represent these correlations with a single number; however, the historical data suggests that we need to include some uncertainty around our estimate at the very least. The equity-bond correlation, in particular, is one of the key estimates used in the asset allocation decisions of an investor but, given its historical variability, it is a relationship which requires careful modelling.

What if advisors were paid based on RULC? A thought experiment

This is more-or-less tongue-in-cheek in case you need to know that before we get too far. I'm just playing around.

What is RULC? Let's start with EDULC first. EDULC is the "expected discounted utility of lifetime consumption." This is the lifecycle econ idea that one might want to plan to optimize aggregate consumption utility over a lifetime and, if one were to randomize and simulate the plan a bazillion (random) lifetimes, the average of the plan over the bazillion is the expected value of the plan. And then we would try to optimize that number by playing with stuff like (primarily) the consumption plan, returns, vol, longevity, etc.  Ignore for now whether this is a valid or appealing thing; assume that it is.  Plan to spend too little and one would under-optimize planned consumption on average. Over-consume and one is likely run out of money at some point and end up consuming at whatever income is available at that time which, again, underoptimizes lifetime utility via the utility penalty for a much reduced level of consumption in the future (ignoring annutization for now which would have a big impact).  Have a good but lumpy plan that is lumpy at the wrong times and one might still have a non-optimal plan due to the convexity  (or is it concavity?) of the utility math.  Also, consider that if a plan is made at day zero, by day 1 things have already changed and vitiated the day-zero plan so that we probably have to re-plan again for day 1. And so on.

34 months of alt risk: small business receivables and term-lending credit

I allocated a small slice of capital to a peer-lending-in-order-to-finance-business-receivables platform in mid 2015. I can't remember if I signed anything about info disclosure so I'll bury the name but it'd probably be easy enough to decode.  This would be a sister strategy to peer to peer consumer lending, by the way, about which I am currently debating an exit.  34 months is a little thin for a track record especially one that had not been tested by a hard business cycle but I thought I'd see what it looks like so far for my own purposes.

Contextualizing my alt-risk strategy in HFR index space

In a recent post here [End of one (strategy) era, the beginning of another] I was trying, with dubious results, to brag about tiny advantages in my private alt risk strategy. This is a strategy that I am leaving behind for a new strategy that is similar but takes way less effort. Here's another look using a different context.

Mark Rzepczynski at Disciplined Systematic Global Macro Views just put out an interesting post (Alternative risk premium indices - Providing insight on what is possible in the ARP space) based on some new indices put out by HFR.  Per Mark:

RH Links - 7/5/18

QUOTE OF THE DAY

Historical data shows that you haven’t died, yet it would be foolish to consider yourself immortal.  Ron Piccinini 

RETIREMENT FINANCE AND PLANNING

Building a Safe Floor of Retirement Income — in Advance. Mike Piper

The solution, rather than buying a bunch of bonds that mature when you turn 70, would be to work on building bond holdings that, when you turn 70, will still have a duration roughly equal to that of the annuity you expect to purchase. This way, the market value of your bonds will rise/fall along with the cost of such an annuity, helping to offset the interest rate risk that you face with the annuity purchase. https://obliviousinvestor.com/building-a-safe-floor-in-advance/

The New Glide Path, Corey Hoffstein

In practice, investors and institutions alike have spending patterns that makes the sequence of market returns a relevant risk factor… For retirees making constant withdrawals, sustained declines in portfolio value represent a significant risk. Trend-following has demonstrated historical success in helping reduce the risk these types of losses… Using backward induction and a number of simplifying assumptions, we generate a glide path based upon investor age and level of wealth. We find that trend-following receives a significant allocation – largely in lieu of equity exposure – for investors early in retirement and whose initial consumption rate closely reflects the 4% level…. we find that trend following commands a significant allocation, particularly in the years and wealth levels where sequence risk is highest, and often is allocated to in lieu of equities themselves.  https://blog.thinknewfound.com/2018/07/the-new-glide-path/

Insights from Life-Cycle Financial Planning, Joe Tomlinson

Research on life-cycle economics and consumption smoothing has a rich history in economics stretching back almost 100 years, and more than a dozen Nobel laureates have contributed to the field of study. Unfortunately, this economics approach has not crossed over to the mainstream financial advice community. But, for the types of issues advisors encounter, it provides both a general unifying theme as well as sensible approaches for particular issues, and therefore deserves more attention. https://www.advisorperspectives.com/articles/2018/07/02/insights-from-life-cycle-financial-planning 

CPI-E and retirement inflation

"From 1982 through December 2012, for example, CPI-E [an experimental Consumer Price Index that captures the spending of adults who are age 62 and above] was 3.07% versus 2.92% for CPI-U. That underscores that retirees ought to remain watchful for signs of inflation and work to preserve their purchasing power from their portfolio withdrawals by embedding inflation hedges into their portfolio holdings."

From How Do Expenses Change in Retirement? by Christine Benz at Morningstar

End of one (strategy) era, the beginning of another

I've been running a self-rolled "alt-risk" strategy that can roughly be labeled something like "multi-asset credit allocation trend following using a systematic rules based approach" for more than seven years.  It actually dates to 2010, if not before, but the quality of my data before 2012 is suspect. I'd say I reliably have about 78 months of coherent data.  For a variety of reasons I am now closing the strategy. In practice it is, in fact, just being rolled into a broader strategy that is less data and time intensive (hey, I'm retired and there is a strong case for simplifying). But before I throw the last shovelful of dirt on the grave of my strategy I wanted to memorialize it for a second. Ignore that none of the strategy's timeframe overlaps any major financial crises.

Thinknewfound on "The New Glidepath"

The New Glide Path, Corey Hoffstein

Excerpts

"In practice, investors and institutions alike have spending patterns that makes the sequence of market returns a relevant risk factor…

For retirees making constant withdrawals, sustained declines in portfolio value represent a significant risk. Trend-following has demonstrated historical success in helping reduce the risk these types of losses…

Using backward induction and a number of simplifying assumptions, we generate a glide path based upon investor age and level of wealth...

We find that trend-following receives a significant allocation – largely in lieu of equity exposure – for investors early in retirement and whose initial consumption rate closely reflects the 4% level…"