Jan 20, 2017

Weekend Links - Jan 20 17

QUOTE OF THE DAY

A [retirement] scheme that does not respond to the performance of the portfolio is likely to underperform one that is responsive to portfolio performance.  - Gordan Irlam 

CHART OF THE DAY


RETIREMENT FINANCE AND PLANNING

What are Longevity Goals? Bob French at Retirement Researcher.  Life expectancy is one of the trickiest parts of financial planning. We can look at actuarial tables and come up with dates when you should pass away with given levels of certainty, but those are nothing but statistical estimates.  Statistics work best for large pension funds and life insurance providers. The group of people those entities cover are so big that they tend to end up looking like the historical averages. They may be off a little bit, but overall, the actuarial tables are probably going to do a pretty good job of predicting what will happen. 

The Opening Game, Dirk Cotton.  the opening of retirement may be the most expensive of the three games…Key risks of the Opening Game include forced retirement, the “Tax Torpedo”, and sequence of returns risk… The Opening Game has its own risks and rewards and decisions made in early retirement can have a dramatic impact on later games. A good retirement strategy will not only consider the impacts of decisions on the Opening Game but also impacts on later games. Plan with the understanding that the three games are different, that each may require its own strategy, that decisions may affect more than the current game, and that you won't know what pieces are still in play until you almost reach the next game.  

[comment: while I think a post like this might actually need a little background and prior knowledge in retirement finance concepts I think that every point made in this post should be required reading for both early and traditional retirees]


Dynamic Programming Methods For Retirement Income, Wade Pfau.  Due to their mathematical complexity, dynamic programming methods are mostly discussed in the realm of academia and have not yet become a common part of the toolkit for individual retirees…Among the other strategies, they [Tomlinson and Irlam] find that the closest match to their optimal dynamic programming solution is the RMD-styled strategy with a fixed 90% stock allocation. Next is a 6.8% fixed percentage strategy with a 90% stock allocation. In third is a constant inflation-adjusted spending strategy with a 60% stock allocation. All of the other asset allocation strategies result in worse outcomes because they lack sufficient aggressiveness.

Why Monte Carlo Analysis StillMatters And The Risk Of A Retiree Black Swan Is [Probably] Overrated, Kitces.com.  So the reality really is that planning for black swans isn’t about getting better planning software to model them: A) because black swans basically don’t exist in annual data; and B) because you wouldn’t really do anything different anyway, because even a properly accounted black swan is not material enough to dramatically change a Monte Carlo analysis or a probability of success. What it’s really about is formulating the plans about how you’re going to respond to a black swan.    

[comment: This article is correct.  A while back I added a feature to my simulator that models unlikely but big spending events. While interesting and important the impact on fail rates is relatively muted.  The more interesting thing, which Kitces points out, is formulating an action plan for what to do (e.g., cut spending by precisely $x) if the sceneario plays out. I've done that a couple times and it is a useful exercise.  I forced myself to figure out exactly what I would need to do to keep spend rates below 4 or 5 % for a while if the stock portion of my portfolio fell 30 or 40% or more for a few years. Then, to take it even further, I went to my income statement expense data and figured out what exactly would have to go and how.  Hard and unpleasant stuff but necessary]

Measuring the Risk of Running Out of Money in Retirement by Grant Gardner, Ph.D.; and Sam Pittman, Ph.D. This paper proposes a simple way to measure sustainability risk—mortality-adjusting—that accounts for the joint occurrence of being alive and running out of money. Different assumptions regarding the age of death of a client leads to very different assessments of retirement sustainability risk. …Planning software and tool providers can help advisers and their clients make more efficient spending and investment decisions by incorporating lifespan uncertainty into their planning tools. The potential errors in sustainability risk assessment caused by using an arbitrarily chosen deterministic planning horizon, and the inefficiency that these errors introduce into retirement decision making, are too large to ignore.  [um, well, I agree] 


Breaking the 4% rule, J.P. Morgan. [marketing sheet by JP Morgan covering a utility optimizing retirement approach. ]  Maximizing expected lifetime utility (i.e., potential derived satisfaction) serves as a more effective benchmark of retirement withdrawal success than typical measures, such as probability of failure. Focusing on utility offers a way to quantify how much satisfaction retirees receive from their portfolio withdrawals. This can help potentially increase investors’ level of income when they are most apt to enjoy their retirement dollars, while still avoiding the risk of premature portfolio depletion.  A dynamic approach to managing withdrawals and asset allocations provides a more effective use of retirement assets than traditional approaches. Adapting to changes in market conditions and investors’ specific situations over time can help maximize the expected lifetime utility generated by retirement assets. This type of dynamic strategy may help provide greater payout consistency and reduce the likelihood of either running out of money or accumulating excess wealth that is unlikely to be used by the investor.   

Portfolio Size Matters, Gordon Irlam.  Journal Of Personal Finance
Volume 13, Issue 2  In contrast to target date funds that vary asset allocation by age alone, it is important to take into account both the client’s age and the client’s portfolio size relative to spending goals when determining an optimal asset allocation. Stochastic Dynamic Programming (SDP) is a mathematical optimization technique that can be used to determine optimal dynamic adjustments to asset allocation in response to evolving portfolio wealth and time horizons. Using SDP, portfolio size appears at least as important to asset allocation decisions as age.  

Competing Risks: Death and Ruin, Dirk Cotton.  Medical research uses methods of analyzing survival studies that are novel in retirement research. We use Kaplan-Meier estimates and competing risks analysis to explore the conditional probability of a retiree outliving her savings as age progresses, the relationship of the competing risks of death and ruin as age progresses, and the timing of portfolio failures due to poor market returns. We find that risk of ruin develops in three stages of a long retirement: a lowrisk period early in retirement with high sensitivity to market returns but few portfolio failures, a middle period in which portfolio failure peaks, and a late period in which death is much more likely than portfolio ruin. 


[ comment: I think this is correct in terms of "three stages."  In my own simulation exercises I see this as well even though I am not explicitly doing Kaplan-Meier (I think I'm doing something vaguely similar but maybe backwards).  Here for example is one output from a longevity-varying simulator.  
I have to be careful when looking at this because the x axis does not represent the age at fail, it represents a "longevity expectation" where the y for a given x is the aggregate fail rate for all wealth-age pairs that have terminated at that terminal age and younger. The fail could have happened earlier.  The point was to be able trace with one's finger along the x, find one's longevity planning expectation and see the fail rate and then maybe go a little further out and check it again and then go all the way to an unrealistic and hyper-conservative max terminal age.  As in Dirk's article, it is easy to see the three stages possibly implied in the chart for this particular example's assumptions: 1) short longevity expectation (58-75) lower fail risk, 2) medium longevity expectation (75-95) rapid rise in ruin estimates, and 3) late expectation (95+) where death risk overtakes ruin risk.   I didn't keep the data but the slope max is probably somewhere around the mode of 87]


MARKETS AND INVESTING 


How Index Funds Democratize Investing, Academic criticismsof indexing lack economic logic and factual support from the real world. WSJ.  The remedies that the papers suggest are also troubling. Some academics propose: first, prohibiting managers of index funds from voting on behalf of shareholders; and second, limiting investment by index funds to one company per sector, thereby eliminating the benefit of diversification that investors have relied on since Mr. Markowitz first published his research more than 50 years ago. 

[really!? How can otherwise smart people be so dumb?] 



One of My Investing Pet Peeves, Ben Carlson.  If you have 40% of your money in a relatively stable asset during stock market losses you don’t really have 90% of your portfolio exposed to risk. 

Asset Allocation is Not for the Faint of Heart,  Adam Butler, ReSolve.  You see, asset allocation is pretty complicated, and not for the faint of heart. For my money, investors have the greatest chance of hitting long-term financial goals by investing in the most diversified portfolio possible (choose your definition), and saving as much as they reasonably can. Any precision beyond this level is a triumph of hope over uncertainty.  

AQR Alternative Thinking 1Q17: Capital Market Assumptions, AQR.  Our current estimate for the long-run real return of U.S. equities is 4.2%, somewhat lower than most other developed markets (average 4.6%) and emerging markets (5.4%). Our current estimate for U.S. 10-year government bonds’ long-run real return is 0.7%. For U.S. investmentgrade and high-yield credit we estimate real returns of 1.4% and 2.1%, respectively. For a riskweighted portfolio of commodities we estimate a long-run real return of around 3%.  From a historical perspective almost all long-only investments have low expected returns today.  

Why 1/N is still a great non-optimal portfolio choice.  None of the fourteen portfolio models consistently dominates 1/N across seven separate datasets (SR and turnover). Victor DeMiguel 

ALTERNATIVE RISK

Diversification: When 1 + 1 < 2? NewFound.  in an environment of high equity valuations and low bond yields, the math for a traditionally allocated balanced portfolio led to a long-term expected return of nearly 0% in real terms… 

Why Some Technical Analysis May No Longer Be Effective: AnInterview With Michael Harris, Forbes.  My opinion based on my own research without making any generalizations is that after prudent risk and money management is added and a trend-following program is diversified, it becomes a smart beta strategy. But that is open to discussion. There are always people who figure out ways of doing things differently and more efficiently and we should not discount this possibility…the key to profitable trading is identifying methods that in turn identify market anomalies before others do. 


SOCIETY AND CAPITAL



The Mad Scientists of Monetary Policy: The War on Cash. Ron Rimkus, CFA Inst. In practice, if the world converted to a cashless society, Mom could still choose what she buys, how she invests, what she does with her money. But she would lose the freedom to withhold her money from the banking system…the ability of the public to choose whether or not to place their money in a bank acts as a vital restraint on the interests of banks, governments, and other actors. The checks and balances of the system are at risk. 

The power and origin of uncertainty shocks. Sr-sv.com.  Uncertainty in economic theory usually means that agents believe that the value of a relevant parameter is a random drawing from a specific probability distribution. An uncertainty shock is a change to that belief in form of a re-assessment of the probability distribution. This can affect any of its moments, such as mean, standard deviation, skewness (bias towards upside or downside) and kurtosis (probability of extreme events). 



No comments:

Post a Comment