Charlie Ellis' 16%

Comments

• JohnChisum

  May 2014

  I'll be interested to see the answer on this one also. I was a victim of old math to new math and back to old math. A whole generation of math illiterates were produced.
• MJG

  May 2014

  Hi Guys,
  
  The little trick to easily reproduce the Ellis calculation is to perform the analysis using dollars and not percentages.
  
  Any dollar amount will do. Let’s assume a 1000 dollar investment.
  
  The active investor was rewarded with a 410 dollar payday; the passive investor’s return was 360 dollars.
  
  The active investor’s cost was 8.50 dollars; the passive investor’s cost was 0.50 dollars.
  
  The excess incremental reward for the active investor over the passive investor was 50 dollars (410-360). The incremental cost for that reward was 8 dollars (8.50-0.50). The incremental cost for the incremental excess return therefore is 8/50 or 16%.
  
  Extra performance is usually a costly matter. A ballplayer who hits at a .320 level draws a much more formidable salary than a .250 hitter. That’s only true if those averages hold water over time. Persistency matters much.
  
  I too admire Charlie Ellis. He is a smart commonsense guy. But in this instance, I believe he is emphasizing the wrong side of the argument. If I could identify a fund manager who generated a lifetime excess return record in the 5% range I would be ecstatic and would gladly pay him 16% of that excess returns. These superinvestors are rare and hard to identify before the fact.
  
  I hope this helps.
  
  Best Regards.
  
• MikeM2

  May 2014

  AHHHHH!
• mrdarcey

  May 2014 edited May 2014

  Another way to put it is that you know what it costs to get the S&P/Total Market performance.
  
  You are paying Mick the Bookie for any outperformance beyond that, IF he or she achieves it. In this case 80bps of 500bps is 16%.
  
  Of course, if Mick the Bookie achieves outperformance, you also have more money to compound.
• glampig

  May 2014

  There are three kinds of people in the world: those who are good at math and those who aren't. Rick
• bee

  May 2014 edited May 2014

  This math (cost differentials) reminds me of a retirement decision I grappled with a few years ago:
  
  "When should I take my retirement pension".
  
  My employer provided me with a retirement pension calculation chart to consider (an annuity chart of sorts) . Calculations started at 25 years of service and maxed out at 37.5 years of service. To simplify the math of this annuity chart, a 59.5 years old worker with 30 years of service could retire with a pension roughly equal to 60% of their salary. An employee could also continuing working (earning a full salary) for 7.5 more years at which time that employee would qualify for a pension roughly equal to 70% of their salary.
  
  What I tried to get across to my follow workers who wanted to continue working towards qualifying for a 70% annuity rate was the fact that they were working for the differential (salary minus a pension @ 60% of last year's salary) or basically 40%.
  
  Of course this worker would earn 40% more money while continuing to work and eventually qualify for a higher pension payment (70%), but from my standpoint these 7.5 years are the "last best" 7.5 years of our lives.
  
  Aside from the enjoyment of work (not really), I was not willing to spend 100% of my time over the next 7.5 years earning a 40% pay differential (salary - pension). Much of this 40% differential disappears from net (take home) income anyway (in the form of pension contributions, union dues, Medicare / SS deduction, state and local tax deductions, and work related expenses).
  
  Cost differential can be steeper than we might image.
  
• cman

  May 2014 edited May 2014

  Putting that 16% in all caps doesn't make it more meaningful. This is the same type of silly argument made with marginal tax rates. It isn't worth getting a million more because you will pay almost half of it in taxes. I don't know any millionaires who thought that way. But saying 50% tax rate makes the point look incredible when arguing ideologically about taxes. Seeing the post here, it seems to work too.
  
  The real argument is that the expense is not tied to a gain over the benchmark. If it was, it would be a no-brainer. You will pay that expense even if you lose money, you will lose more money because of the higher expense. The marginal rate for some hypothetical gain is irrelevant and meaningless. You can make it 50% by making the manager over perform by just 1.6%.
  
  Ellis is being generous by arbitrarily picking just 16% but also using a sensationalist type of argument even if the point being made about mutual fund costs is valid.
• Vert

  May 2014 edited May 2014

  I suppose you could say it's the mathematical equivalent of idle rhetoric. It might not be false but it doesn't mean very much. I might say that the costs for incremental return on index funds are infinite. Say the costs of the index fund is .001. The incremental return will be 0/.001 = infinity.
  
  Or if your friend gave you a few stocks that beat an index without charging you anything then calculating the costs of that incremental return will blow up the entire system since you can't divide by 0.
  
  You probably should meet these oddities with a shrug of the shoulders.