Making Sense Out Of: Fundamentals of Performance Measurement +

Download as pdf or txt
Download as pdf or txt
You are on page 1of 57

5/19/2020

Fundamentals of Performance  What we’ll cover: RORs


Measurement + • Returns that don’t make sense
• A focus look at the IRR
• How to calculate gross- and net-of-
fee returns
• Calculating subportfolio returns
• Handling income accruals
• Different approaches to cash flows
Module #2: Rates of Return, Part II + Benchmarks
David D. Spaulding, DPS, CIPM
Copyright © The Spaulding Group, Inc. 2020 1 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 2 2

What we’ll cover: Benchmarks Making sense


out of
• A review of commonly ref’d criteria
• Different approaches to market indexes
• Impact of turnover
• Benefits/shortcomings of peer groups
• Review absolute benchmarks
• A look at custom benchmarks
Uncovering the reasons
behind problem returns
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 3 3 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4 4

1
5/19/2020

A return doesn’t look right
What do you do?

Sometimes, 
• returns look wrong, but they’re right
• returns look right, but they’re wrong
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 5 5 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 6 6

We may be encountering a paradox A paradox is
“a situation or proposition that seems to be absurd or 
contradictory but is or may be true.” 

Source: Encarta World English Dictionary by way of How Mathematicians Think, by William Byers

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 7 7 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 8 8

2
5/19/2020

Has the market been unusually volatile?
Begin with the data:

• Valuations correct?
• Accurately capture the cash 
flows?
• (values & flow dates?) 2020 will bring with it weird situations
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 9 9 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 10 10

Beware of
“negative space” Follow the
money

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 11 11 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 12 12

3
5/19/2020

What’s your cash flow treatment?
• We often see problems with the timing of the cash 
flows.
• If all flows are treated as start‐ or end‐of‐day, problems 
can arise
• We generally recommend start for inflows, and end‐of‐
day for outflows
• Check your timing, and see if this might be the cause
• It might be appropriate to revise your cash flow timing 
policy
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 13 13 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 14 14

Not just external flows! What was the timing of the client’s flows?
• Internal cash flows can cause nonsensical (and  • Did the client make contributions as the market was 
arguably incorrect) sub‐portfolio returns rising, before a dip occurred?
• Were inflows relatively large?
• Some investors expect markets, that have been rising, 
to continue to rise, and so continue to add money
• When there’s a shift in direction, the resulting returns 
can look invalid, when they’re actually correct

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 15 15 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 1616

4
5/19/2020

Geometric linking can also cause issues: If,…
Need to be aware of its limitations! • The data is clean
• Your cash flow timing 
is appropriate
• The market has seen 
some volatility
Then it’s likely you’ll 
need to justify the 
returns, as they’re 
probably correct. 
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 17 17 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 1818

We’ll consider three examples of paradoxical situations Consider this scenario; 
a bit extreme, but it makes a point
Involves an external cash flow, that 
results in questionable results Activity Position Value
Purchase 1,000 
January 1 STAR @ $1/
1,000 STAR @ $1 $1,000
Involves internal cash flows, that  June 30 1,000 STAR @ $100 $100,000
yield nonsensical asset class returns $1,000,000 inflow; 
1,000 STAR @ $100 
July 1 Purchase 1,000 
1,000 DOG @ $1,000
$1,100,000
Demonstrates how linking in  DOG @ $1,000/
1,000 STAR @ $100 
“negative space” can cause problems December 31 1,000 DOG @ $1,00
$200,000

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 19 19 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2020

5
5/19/2020

Situation summary: Question:
• Client began the year with a $1,000 investment Do you think the return for the year will be
• For first half of the year, investment grew to $100,000 • Positive, or
• Client added $1 million, bringing total invested  • Negative?
amount to $1,001,000 Do you think the result will be a
• At the end of the year, investment dropped to  • Big number, or a
$200,000, meaning a $801,000 unrealized loss • Small number?

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 2121 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2222

Let’s calculate the return
Our solution might surprise you!
• January 1 – June 30:
VE 100,000
R1  1  1  9,900%
V0 1,000
• July 1‐December 31:
VE 200,000
1,718.18%
R2  1  1  8182%
.
V0 , ,000
1100

• January 1 – December 31:
n
R    Ri  1  1   R1  1 R2  1  1 
i 1
Return = 1718.18%
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2323 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2424

6
5/19/2020

Recap: Interestingly …
• We started with $1,000, which went to $100,000
The money‐weighted (IRR) for the year is ‐96%
• We added $1 million, and ended with $200,000
The year’s return is 1,718.18%. Seriously?
Doesn’t this make a lot more sense?
The net loss for the year is $801,000!
How could we explain this to our client?
How would you explain it?

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 2525 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2626

Reversing the order of the flows:
• Client begins the year with a $1,000,000 investment: 
1 million shares of STAR @ $1 per share
• For first half of the year, investment grew to $100MM
^ • Client sells 999,000 shares of STAR; withdraws $98.9 
million, and buys $1,000 DOG @ $1,000/share
• At the end of the year, investment dropped to 
$200,000, meaning a sizable net gain for the year

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 27 27 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2828

7
5/19/2020

Let’s calculate the return Chances are, this surprises you, too!


• January 1 – June 30:
100,000,000
R1   1  9,900%
1,000,000
• July 1‐December 31:
200,000
1,718.18%
R2   1  8182%
.
1100
, ,000
• January 1 – December 31:
n
R    Ri  1  1   R1  1 R2  1  1 
i 1
Return = 1718.18%
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 2929 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3030

The results are identical!  They’re the same because the portfolio manager 
But how can this be, since: performed identically in both instances
• The two portfolios were invested in the same security during 
• In the first scenario, the client lost $801,000 the first half of the year; the only difference was the amount: 
• And in the second, the client made over $98 million $100,000 vs. $1 million
• How can we get the same results? • The return has to be the same for the first half of the year
• During the second half, they had identical portfolios
• The return has to be the same for the second half of the year
• We use time‐weighting to eliminate or reduce the impact of 
client cash flow decisions
• The driver: those cash flow decisions
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3131 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3232

8
5/19/2020

Do these situations occur? What do you do when these occur?
• Absolutely; granted, not to this level, but they do • Follow the money!
• If you’re in performance measurement long enough, you’ll run  • Look for the cash flows.
into situations where: • What happened before and after the flow(s).
The client loses money, but has a positive return • Chances are, there were inflows during the period when the 
• When this occurs, there will almost always be an inflow prior to  portfolio’s return was going up
the market downturn • And then there was a drop in the market, resulting in a loss
• The return is right: it’s measuring how the manager did • However, overall, the portfolio did well for the period
• But it won’t make sense to the client, without some help • Someone will have to explain to the client that their decision to 
• By also reporting the IRR, you will provide the client with a  add money when they did was the main cause of the loss: “buy 
meaningful number for his/her return (reflecting the flow(s) high/sell low”!!!
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3333 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3434

Can the opposite occur? An internal cash flow dilemma
• That is:
The client makes money, but has a negative return?
• Yes, but it seems to be less likely
• Research shows that most investors chase returns

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 3535 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3636

9
5/19/2020

An internal cash flow dilemma Here’s the year’s activity
Equities Bonds Cash Total
• Consider this situation: BMV
ROR
10,000.00
-2.00%
10,000.00
-1.50%
80,000.00
0.80%
100,000.00
0.29%
Equities Bonds Cash Total EMV 1Q 9,800.00 9,850.00 80,640.00 100,290.00
ROR 2.32% -0.85% 3.24% 6.01% ROR -3.00% -2.00% 0.80% 0.15%
EMV 2Q 9,506.00 9,653.00 81,285.12 100,444.12
• Our portfolio’s return is greater than any of its parts Rebalance
BMV 60,000.00 30,000.00 10,444.12 100,444.12
• Why? ROR 3.50% 1.20% 0.80% 2.53%
EMV 3Q 62,100.00 30,360.00 10,527.67 102,987.67
ROR 4.00% 1.50% 0.80% 2.94%
EMV 4Q 64,584.00 30,815.40 10,611.89 106,011.29
Equities Bonds Cash Total
ROR 2.32% -0.85% 3.24% 6.01%

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 3737 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 3838

What about the IRR? Do they make more sense? Consider this actual situation
Equities Bonds Cash Total
MWRR 11.84% 2.33% 3.24% 6.01%

 50,494   64,584 
IRREquities  10,000     0
 1  r  1/ 2   1  r 
 20,347   30,815.40 
IRRBonds  10,000     0
 1  r  1/ 2   1  r 
  70,841
IRRCash  80,000     10,66189
. 0
 1  r  1/ 2 
Since the manager controls internal flows, IRR is a better performance measure

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 3939 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4040

10
5/19/2020

Situation provided by a client:  It turns out, this is like being in a black hole
MTD 3/26
Portfolio
Portfolio
MTD 3/26 Day 3/27 MTD 3/27
‐28.38%
‐28.38% 6.02% ‐24.07%
“Negative space”
Benchmark
Benchmark ‐24.48%
‐24.48% 5.74% ‐20.15%
Excess Return
Excess Return ‐3.90%
‐3.90% 0.28% ‐3.92%
• The portfolio’s MTD excess return through 3/26 is ‐3.90%
• On 3/27, the portfolio outperforms benchmark by 0.28%
• So, what would you expect? An improved MTD excess return, right?
• However, when this day’s return is linked to the running monthly 
return, the excess return gets worse, not better
• The manager, understandably, was incredulous about this
How could I be worse off, given my prior day’s success?
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4141 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4242

First, returns do not compound the same in  Notice how the impact of the 0.28% alpha varies:
Negative Space Positive Space

“negative space” as they do in “positive space” Panel 1
Portfolio
MTD 3/26 Day 3/27
‐5.90% 6.02%
MTD 3/27
‐0.24%
MTD Δ
5.66%
Panel 6
Portfolio
MTD 3/26 Day 3/27
2.00% 6.02%
MTD 3/27
8.14%
MTD Δ
6.14%
Benchmark ‐2.00% 5.74% 3.63% 5.63% Benchmark 5.90% 5.74% 11.98% 6.08%
Excess Return ‐3.90% 0.28% ‐3.86% 0.04% Excess Return ‐3.90% 0.28% ‐3.84% 0.06%

Second, the impact a return will have when  Panel 2
Portfolio
MTD 3/26
‐18.90%
Day 3/27
6.02%
MTD 3/27
‐14.02%
MTD Δ
4.88%
Panel 7
Portfolio
MTD 3/26
15.00%
Day 3/27
6.02%
MTD 3/27
21.92%
MTD Δ
6.92%

compounding will vary, depending on the size of  Benchmark
Excess Return
‐15.00%
‐3.90%
5.74%
0.28%
‐10.12%
‐3.90%
4.88%
0.00%
Benchmark
Excess Return
18.90%
‐3.90%
5.74%
0.28%
25.72%
‐3.80%
6.82%
0.10%

Panel 3 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ Panel 8 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ

the linked return Portfolio
Benchmark
‐28.38%
‐24.48%
6.02%
5.74%
‐24.07%
‐20.15%
4.31%
4.33%
Portfolio
Benchmark
24.48%
28.38%
6.02%
5.74%
31.97%
35.75%
7.49%
7.37%
Excess Return ‐3.90% 0.28% ‐3.92% ‐0.02% Excess Return ‐3.90% 0.28% ‐3.78% 0.12%

Panel 4 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ Panel 9 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ
Portfolio ‐53.90% 6.02% ‐51.12% 2.78% Portfolio 50.00% 6.02% 59.03% 9.03%
Benchmark ‐50.00% 5.74% ‐47.13% 2.87% Benchmark 53.90% 5.74% 62.73% 8.83%
Excess Return ‐3.90% 0.28% ‐3.99% ‐0.09% Excess Return ‐3.90% 0.28% ‐3.70% 0.20%
Panel 5 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ Panel 10 MTD 3/26 Day 3/27 MTD 3/27 MTD Δ
Portfolio ‐93.90% 6.02% ‐93.53% 0.37% Portfolio 96.00% 6.02% 107.80% 11.80%
Benchmark ‐90.00% 5.74% ‐89.43% 0.57% Benchmark 99.90% 5.74% 111.37% 11.47%
Excess Return ‐3.90% 0.28% ‐4.11% ‐0.21% Excess Return ‐3.90% 0.28% ‐3.58% 0.32%

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 4343 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4444

11
5/19/2020

The variance/impact is actually linear across returns The explanation
MTD 3/26 Day 3/27
MTD 3/26 MTD 3/27 MTD Δ
Portfolio
Portfolio ‐28.38%6.02%
‐28.38% ‐24.07% 4.31%
Relationship Between Size of Cumulative Return & Incremental Change
14% Benchmark ‐24.48%
Benchmark ‐24.48%5.74% ‐20.15% 4.33%
12%
Excess Return‐3.90%
Excess Return ‐3.90% 0.28% ‐3.92% ‐0.02%
10% • While the manager’s return was superior to the benchmark’s, 

Incremental Change

8% Since its MTD return was so very low relative to the benchmark, 
6% its impact wasn’t as great (4.31%) as the benchmark’s (4.33%)
4% • The impact of a return will vary, depending on size
2% • It is not a uniform change; but rather increases as returns 
‐100% ‐80% ‐60% ‐40% ‐20%
0%
0% 20% 40% 60% 80% 100%
increase
Cumulative Return

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 4545 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 4646

Before we close out What do you make of this scenario?


this section, let’s MONTH END 
DATE
START 
VALUE
END VALUE NET ADDITIONS NET GAINS
MONTHLY 
RETURN 
CUMULATIVE 
RETURN GROSS

consider the
GROSS
7/31/2017 $0 $177,577 $176,038 $1,539 0.86% 0.86%
8/31/2017 $177,577 $176,942 ($47) ($589) ‐0.33% 0.53%

following …
9/30/2017 $176,942 $180,111 ($2,183) $5,352 3.04% 3.59%
10/31/2017 $180,111 $181,111 ($318) $1,318 0.73% 4.35%
11/30/2017 $181,111 $184,858 $1,820 $1,928 1.06% 5.45%
12/31/2017 $184,858 $247,825 $61,011 $1,956 1.12% 6.63%
1/31/2018 $247,825 $473,769 $217,808 $8,136 3.97% 10.86%
2/28/2018 $473,769 $508,316 $65,845 ($31,298) ‐5.80% 4.43%
3/31/2018 $508,316 $579,800 $77,051 ($5,567) ‐0.91% 3.48%
$176,038 $579,800 $420,987 ($17,226)

What steps might you take?


What do you tell your client (who is a STEM-type person)?

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 4747 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 4848

12
5/19/2020

What did I do? Is this


MONTH END  START 
MONTHLY 
CUMULATIVE 
clear
enough
END VALUE NET ADDITIONS NET GAINS RETURN 
DATE VALUE RETURN GROSS
GROSS
7/31/2017 $0 $177,577 $176,038 $1,539 0.86% 0.86%
8/31/2017
9/30/2017
$177,577
$176,942
$176,942
$180,111
($47)
($2,183)
($589)
$5,352
‐0.33%
3.04%
0.53%
3.59% for
your
10/31/2017 $180,111 $181,111 ($318) $1,318 0.73% 4.35%
11/30/2017 $181,111 $184,858 $1,820 $1,928 1.06% 5.45%
12/31/2017 $184,858 $247,825 $61,011 $1,956 1.12% 6.63%
1/31/2018
2/28/2018
$247,825
$473,769
$473,769
$508,316
$217,808
$65,845
$8,136
($31,298)
3.97%
‐5.80%
10.86%
4.43% client?
3/31/2018 $508,316 $579,800 $77,051 ($5,567) ‐0.91% 3.48%
$176,038 $579,800 $420,987 ($17,226)

I calculated the IRR (using the XIRR function) and got


-6.62% (which happens to equal the Mod Dietz result)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 4949 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5050

Let’s find a Solve the following, using the exact method


way to solve  VE
R    i

  1
 V0i
this puzzle! 
Investor #1
1/1/2017 Start          100,000
End of Pd 1          120,000
7/1/2017 Addition       1,000,000
Start Pd 2       1,120,000
12/31/2017 End of Pd 2          952,000
Net gain/loss         (148,000)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5151 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5252

13
5/19/2020

Now, solve this one, also by using the Our TWRR results
exact method V 
R      1 Ei
1/1/2017 Start
Investor #1
         100,000
 VE 
R#1    i   1
 V0i 
 V0i  End of Pd 1
7/1/2017 Addition
         120,000
      1,000,000
20.00%
 120,000 952,000 
   1
Start Pd 2       1,120,000  100,000 1120
, ,000 
Investor #2 12/31/2017 End of Pd 2
Net gain/loss
         952,000
        (148,000)
‐15.00%
2.00%  2.00%
1/1/2017 Start       1,000,000
End of Pd 1       1,200,000 Investor #2  VE 
1/1/2017 Start       1,000,000 R#2    i   1
7/1/2017 Withdrawal         (900,000) End of Pd 1       1,200,000 20.00%  V0i 
7/1/2017 Withdrawal         (900,000)  1,200,000 255,000 
Start Pd 2          300,000 Start Pd 2          300,000    1
12/31/2017 End of Pd 2          255,000 ‐15.00%
 1,000,000 300,000 
12/31/2017 End of Pd 2          255,000 Net gain/loss          155,000 2.00%  2.00%
Net gain/loss          155,000
Are the same! How would you explain this?
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5353 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5454

Now, use Mod Dietz across the Our Money-weighted (MD) results
full year as a proxy for the IRR 1/1/2017
Investor #1
Start          100,000 R MD#1 
VE  V0   C
V0   WC
(assume mid-point flow) End of Pd 1
7/1/2017 Addition
         120,000
      1,000,000 952,000  100,000  1,000,000
Start Pd 2       1,120,000 
Investor #1 Investor #2
12/31/2017 End of Pd 2          952,000 1,000  (0.5)(1,000,000)
1/1/2017 Start          100,000 1/1/2017 Start       1,000,000
End of Pd 1          120,000 End of Pd 1       1,200,000
Net gain/loss         (148,000)  24.67%
7/1/2017 Addition       1,000,000 7/1/2017 Withdrawal         (900,000)
Investor #2
Start Pd 2       1,120,000 Start Pd 2          300,000 VE  V0   C
1/1/2017 Start       1,000,000 R MD# 2 
V0   WC
12/31/2017 End of Pd 2          952,000 12/31/2017 End of Pd 2          255,000
Net gain/loss         (148,000) Net gain/loss          155,000 End of Pd 1       1,200,000
7/1/2017 Withdrawal         (900,000) 255,000  1,000,000  ( 900,000)

Start Pd 2          300,000 1000,,000  (0.5)( 900,000)
12/31/2017 End of Pd 2          255,000  2818%
.
Net gain/loss          155,000
IRR#1=-23.33% IRR#2=27.07%
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5555 Copyright © The Spaulding Group, Inc. 2020 56

14
5/19/2020

Our Money-weighted (MD) results


Investor #1 VE  V0   C
Questions?
1/1/2017 Start          100,000 R MD#1 
End of Pd 1          120,000 V0   WC
7/1/2017 Addition       1,000,000 952,000  100,000  1,000,000

What do you tell


Start Pd 2       1,120,000 
12/31/2017 End of Pd 2          952,000 1,000  (0.5)(1,000,000)
Net gain/loss         (148,000)  24.67%

your client?
Investor #2
VE  V0   C
1/1/2017 Start       1,000,000 R MD# 2 
End of Pd 1       1,200,000 V0   WC
7/1/2017 Withdrawal         (900,000) 255,000  1,000,000  ( 900,000)

Start Pd 2          300,000 1000,,000  (0.5)( 900,000)
12/31/2017 End of Pd 2          255,000  2818%
.
Net gain/loss          155,000

IRR#1=-23.33% IRR#2=27.07%
Copyright © The Spaulding Group, Inc. 2020 57 Copyright © The Spaulding Group, Inc. 2020 58

Hopefully you’ll agree: And so, let’s spend a bit


There are times when more time on the IRR
the IRR makes sense!

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 5959 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6060

15
5/19/2020

1.We’ll discuss different approaches


to solve the IRR
2.Review the Excel XIRR function
3.Discuss issues surrounding the IRR
Whichever method we employ, we repeat the process
until a sufficiently accurate value is reached
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6161 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6262

3 Major
Newton-Raphson
Methods method, attributed to
Isaac Newton &
Joseph Raphson
∙ Newton Raphson
∙ Secant
∙ Bisection

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6363 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6464

16
5/19/2020

A Demonstration of the
Newton-Raphson
Method

Source: Wikipedia

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6565 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6666

The secant method can be thought of as a


finite difference approximation of Newton's A Demonstration of the
method. However, the method was developed Secant Method
independently of Newton's method, and First 
predates it by over 3,000 years. estimate
Final 
estimate

Source: Wikipedia

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6767 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6868

17
5/19/2020

The bisection method repeatedly bisects an


interval, and then selects a subinterval in which the Where to start?
solution must lie, for further testing.
• Recall that the IRR is an 
iterative method.
• We need a starting point.
• Do we pick a number at 
random?

Solution

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 6969 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7070

Let’s walk through an


example, using bisection

We typically start with the  VB = 100,000 (12/31/14)


Modified Dietz method C1 = +15,000 (2/7/15)
It’s generally considered the “first  C2 = +25,000 (4/15/15)
approximation” for the IRR VE = 150,000 (6/30/15)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7171 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7272

18
5/19/2020

We begin with Modified Dietz Recall the IRR formula from earlier
D W
Ci VE
12/31/14 VB 100,000 0  VB   
2/7/15 C1 15,000 38 0.80
(1  r )ti (1  r )
4/15/15 C2 25,000 105 0.43 The ti values are the weights, that represent the time
6/30/15 VE 150,000
CD 181
remaining from the flow date; i.e., it’s 1-(CD-D)/CD
Modified Dietz 8.16%
D W t (IRR)
CD  Di  1 CD  Di  1 181  38  1 12/31/14 VB 100,000
Wi  W1    0.80
CD CD 181 2/7/15 C1 15,000 38 0.80 0.21
VE  VB   Ci 150,000  100,000  (15,000  25,000 ) 4/15/15 C2 25,000 105 0.43 0.58
R MD   6/30/15 VE 150,000
VB   Wi Ci 100,000  ( 0.80  15,000  0.43  25,000 )
 816%
. CD 181

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7373 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7474

We now employ bisection, using Modified Now, let’s try the XIRR function
Dietz as our first guess for the solution
Ci VE
0  VB   
(1  r )ti (1  r )
IRR Guess #1 8.16% ‐42.06
Bisection 
Method

Guess #2 8.25% 60.65


Guess #3 8.20% 4.44
Guess #4 8.19% ‐6.80

Our solution is 8.20%. While the result isn’t 0, it’s reasonably close.
Note that 8.19% and 8.20% border zero; and since 8.20%’s result
(4.44%) is closer, that’s our solution.
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7575 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7676

19
5/19/2020

XIRR function format: Note that the signs of the values are key:
• Inflows carry a minus sign
=xirr(values,dates,[guess])
• Outflows carry a positive sign
• Values = beginning, ending, and cash flows
• Dates = dates associated with the values Values (IRR)
12/31/14 VB (100,000)
• Guess is the first estimate, which is optional 2/7/15 C1 (15,000)
4/15/15 C2 (25,000)
6/30/15 VE 150,000

XIRR = 17.22%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7777 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 78 78

Why might it surprise you? Chances are, you won’t want to solve for
XIRR = 17.22% the IRR manually …
That’s not even close to our bisection
result (8.19%)! Let your software
do the work
The XIRR function annualizes the result,
so we need to de-annualize it.
XIRRDe  annualized  (1  r ) ( Days / 365)  1
 (1  17.22%) (181/ 365)  1  8.20%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 7979 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8080

20
5/19/2020

A few points about the IRR
Let’s review one more
• Must solve iteratively; various
approaches; main ones: ROR formula:
• Newton/Raphson Modified BAI (aka,
• Secant
• Bisection
Linked IRR)
• Issue with multiple solutions
• Don’t believe relevant for
investing
• Issue with no solution
• Don’t believe relevant for
investing
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8181 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8282

As mentioned yesterday,  We’ve already


in 1968, the BAI (Bank  addressed the exact
Administration Institute)  method.
came up with three 
approaches to time‐ The “linked IRR” is the
weighting: same as the Modified
• Exact BAI
• Linked IRR
• Regression (no one uses)
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8383 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8484

21
5/19/2020

We know that the IRR is a The result will equal


money-weighted method.
or be very close to
However, by subdividing the what we get with the
period (e.g., a year into Modified Dietz
months) and calculating the method.
IRR for each sub-period and
linking, the result is a time-
weighted return.

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8585 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8686

Contrasting IRR, BAI Exact, Modified 
If you’re going to use an
approximation method, why
Dietz, and Modified BAI/Linked IRR
use Modified BAI, when Value Weight
IRR
Weight Weight Value
Modified Dietz yields
M1 M2 CF M3 M4 M5 M6 CF LCF M7 Mn
essentially the same result,
and is a much simpler
Value
Value

Value

Value

Value
Value
Value

Value
Value
Value

Value

formula to employ?
Exact
Value

Value
Value

Value
Value

Mod Dietz / Mod BAI

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8787 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8888

22
5/19/2020

Any Before we
conclude this
Questions? section …
Let’s talk a bit
more about
Modified Dietz

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 8989 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9090

Modified Dietz:
Time‐weighted or Money‐weighted?
How can this be?
• Answer: both! (confused?) • When Modified Dietz stands alone, by itself, 
• Modified Dietz actually approximates  without any linking, it’s technically a money‐
both! weighted return
Time
Weighted
Modified
Money
Weighted
• It’s typically used as the first approximation for the 
Dietz
(approximates) (approximates) IRR (the true, money‐weighted ROR)

Unless
cash flows
are large!

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 91 91 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9292

23
5/19/2020

How can this be? Contrasting Mod Dietz as TWRR / MWRR
Chain‐linking of monthly returns 
• When we geometrically link it, it’s an  Approximates the True, Exact, TWRR
approximation for time‐weighting
• Analogous to the linked IRR (aka Modified IRR  Mod DJan Mod DFeb Mod DMar Mod DApr Mod DSep Mod DOct Mod DNov Mod DDec

and Modified BAI)
. . .

ModD Year

Approximates the True, Exact, MWRR (IRR)
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9393 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9494

We saw previously how the Modified Dietz is  Modified Dietz has the same problem when 
impacted by cash flows for time‐weighting approximating the IRR
Case #1: No cash flow Case #2: 5% cash flow
Portfolio
ROR BMV EMV
Portfolio
ROR BMV EMV
• Length of period: 90 days
• Beginning market value: $99,450
1st half 1% $ 1,000.00 $ 1,010.00 1st half 1% $ 1,000.00 $ 1,010.00
2nd half 5% $ 1,010.00 $ 1,060.50 2nd half 5% $ 1,060.50 $ 1,113.53
Modified Dietz Modified Dietz

• Ending market value: $343,610
C= $ - C= $ 50.50
equals the True approximates the
Mod D= 6.05% Mod D= 6.15%
True = 6.05% Time-weighted True = 6.05% True Time-
Difference = 0.00% ROR Difference = 0.10% weighted ROR
Case #3: 10% cash flow Case #4: 25% cash flow • Cash flow: $500,000
Portfolio Portfolio

1st half
ROR
1%
BMV
$ 1,000.00
EMV
$ 1,010.00 1st half
ROR
1%
BMV
$ 1,000.00
EMV
$ 1,010.00 • Date of cash flow: 62nd day.
2nd half 5% $ 1,111.00 $ 1,166.55 2nd half 5% $ 1,262.50 $ 1,325.63

C= $ 101.00 Modified Dietz's C= $ 252.50 Modified Dietz's


Mod D= 6.24% error is beginning Mod D= 6.49% approximation is
True = 6.05% to grow True = 6.05% unacceptable
Difference = 0.19% Difference = 0.44%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9595 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9696

24
5/19/2020

Solving … ‐100.33%
Nice return: ‐100.33%
• So, how did we manage to lose more than 100% of our 
VE  VB  C money? 
R
VB  W  C • By my calculations, we had a total investment of 
$599,450 ($99,450 + $500,000) and ended with 
CD  D 90  62
W   0.31111 $343,610, meaning we lost $255,840. 
CD 90
• Hardly a 100%+ loss.
343,610  99,450  500,000
R   10033
.   100.33%
99,450  0.31111  500,000
Copyright © The Spaulding Group, Inc. 2020 97
Copyright © The Spaulding Group, Inc. 2020 97 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9898

IRR yields a return of ‐76.19% Conclusion • As with using the Modified 


Dietz to approximate the 
Make more sense? Guess Result true, time‐weighted ROR
-100.33% #NUM!

Ci EMV
-99% 22,328,600.07 • It falls short when 
-80% 103,354.65
0   BMV   approximating the IRR if 
 r  1 t i
 r  1 -60% (180,379.68)
-70%
-75%
(100,059.91)
(24,349.91)
there are very large cash 
-77% 18,346.55 flows, too
-76% (4,139.98)
-76.50%
-76.25%
6,798.28
1,255.60
• Refer to our on‐line 
-76.10% (1,999.07) newsletter (March 2008) 
-76.20% 164.88
-76.15% (920.00) for more details on this.
-76.18% (269.77)
-76.19% (52.56)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 9999 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 100
100

25
5/19/2020

Calculating gross- and net-of-fees


• Gross-of-fee returns means that the return is
gross of the advisory fee, meaning that the fee’s
impact has not been reflected. The return will be
reduced by trading expenses.
• Net-of-fee returns means that the return is net
of the advisory fee, meaning that the fee’s
impact has been reflected. The return is reduced
by both trading expenses and advisory fees.

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 101
101 Copyright © The Spaulding Group, Inc. 2020 102

How are fees charged? Calculating the fee if taken from


• If the fees are taken from the corpus of the the corpus of the account
account, it means that the fees have come from
• Need to be able to identify the fee transaction
the account itself; that the portfolio’s market
• Gross of fee return: treat the fee as a cash flow
value has been reduced because of the fees.
(a withdrawal)
• Alternatively, the client may pay fees from
• Net of fee return: ignore the fee
another source, in which case the portfolio
doesn’t reflect the fee at all.

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 103
103 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 104
104

26
5/19/2020

Example of calculating the GOF if Example of calculating the NOF if


taken from the corpus of the account taken from the corpus of the account
• BMV = $100,000 • BMV = $100,000
• Fee = $250; taken on 15th June • Fee = $250; taken on 15th June
• Value at start-of-day on 15th = 100,500 • Value at start-of-day on 15th = 100,500
• EMV = $100,750 • EMV = $100,750
n V Ei 100,500 100,750 n V Ei 100,750
RORGOF / TD   1   1  100%
. ROR NOF / TD   1  1  0.75%
i 1 V0i 100,000 100,250 i 1 V0i 100,000

VE  V0  C 100,750  100,000  ( 250) VE  V0  C 100,750  100,000


RORGOF / MD    100%
. ROR NOF / MD    0.75%
V0  WC 100,000  0.5  ( 250) V0  WC 100,000

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 105
105 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 106
106

Calculating the fee if taken from Example of calculating the GOF if


outside the account taken from outside of the account
• BMV = $100,000
• Need to be able to identify the fee amount and • Fee = $250; treat as a contribution
date • Value at start-of-day on 15th = 100,500
• Gross of fee return: ignore the fee • EMV = $101,000
• Net of fee return: treat the fee as a n V Ei 101,000
RORGOF / TD   1  100%
contribution, with no impact on the market values i 1 V0i 100,000
.

V E  V0  C 101,000  100,000
RORGOF / MD    100%
.
V0  WC 100,000

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 107
107 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 108
108

27
5/19/2020

Example of calculating the NOF if


taken from outside of the account Should you
• BMV = $100,000 accrue
• Fee = $250; treat as a contribution
• Value at start-of-day on 15th = 100,500
advisory
• EMV = $101,000 fees?
n V Ei 100,500 101,000
ROR NOF / TD   1   1  0.75%
i 1 V0i 100,000 100,500  250

VE  V0  C 101,000  100,000  250


ROR NOF / MD    0.75%
VB  WC 100,000  0.5  250

Copyright © The Spaulding Group, Inc. 2020 109 Copyright©©The


Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 110
110

Scenario #1

GIPS recommends that Jan


GOF
1.79%
Accrual
1.71%
Non‐Accrual
1.79%
compliant firms accrue fees Feb
Mar
3.72%
‐0.04%
3.64%
‐0.12% 5.28%
3.72%
‐0.29% 5.27%

Why?
Apr 0.91% 0.83% 0.91%
May 1.16% 1.08% 1.16%
Jun 0.48% 0.40% 2.32% 0.23% 2.32%
Because it’s supposed to be Jul 1.93% 1.85% 1.93%
Aug 0.05% ‐0.03% 0.05%
more accurate! Sep 1.93% 1.85% 3.69% 1.68% 3.70%
Oct 2.22% 2.14% 2.22%
But, is it? Let’s see … Nov
Dec
2.81%
0.98%
2.73%
0.90% 5.86%
2.81%
0.73% 5.86%
Year 19.41% 18.24% 18.24%
Used S&P 500returns for 2017

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 111
111 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 112
112

28
5/19/2020

Scenario #2

Jan
GOF
‐5.07%
Accrual
‐5.15%
Non‐Accrual
‐5.07%
And so, it appears that there
Feb ‐0.41% ‐0.49% ‐0.41% is minimal differences (if
Mar 6.60% 6.52% 0.53% 6.35% 0.55%
Apr 0.27% 0.19% 0.27% any) between accruing and
not accruing.
May 1.53% 1.45% 1.53%
Jun 0.09% 0.01% 1.64% ‐0.16% 1.64%
Jul 3.56% 3.48% 3.56%
Aug
Sep
‐0.12%
‐0.12%
‐0.20%
‐0.20% 3.06%
‐0.12%
‐0.37% 3.05%
Is there any other reason
Oct ‐1.94% ‐2.02% ‐1.94% firms might want to accrue?
Nov 3.42% 3.34% 3.42%
Dec 1.82% 1.74% 3.01% 1.57% 3.01%
Year 9.55% 8.47% 8.48%
Used S&P 500 returns for 2016

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 113
113 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 114
114

Calculating 
Sub‐Portfolio Rates of Return
• Asset Class Level
• Industry Level
• Sector Level
• Sub‐sector Level
• Security Level

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 115
115 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 116
116

29
5/19/2020

Calculating the returns
What are cash flows for sub‐portfolios? Things to consider
• Purchases = Inflows • Which to use?
• Sales = Outflows • Time‐weighted, or
• Income = Outflows • Money‐weighted
• Question: Who controls these (internal) cash flows, the 
• Corporate Actions = both 
client or the manager?
inflows & outflows
• If we do TWRR, since the flows tend to be large (> 10%), 
• External Cash flows = both we would use an exact method (revalue when flows 
occur).
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 117
117 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 118
118

Exercise: Calculate the exact TWRR for Solution


the following scenario Vt 1 1,000  5 110,000 5,000 110,000
R 1   1    1  0.00%
# Shares Price Value Vt 10,000 11,000  5 10,000 55,000
6/1 Starting position       1,000 $10 $   10,000
# Shares Price Value
6/15 Price drops to $5, and so we buy more 6/1 Starting position       1,000 $10 $   10,000
6/15 Purchase     10,000 $5 $   50,000 6/15 Price drops to $5, and so we buy more
6/15 Purchase     10,000 $5 $   50,000
6/30 Price returns to starting price
6/30 Price returns to starting price
6/30 Ending position     11,000 $10 $ 110,000 6/30 Ending position     11,000 $10 $ 110,000
Unrealized gain     10,000 $5 $   50,000 Unrealized gain     10,000 $5 $   50,000
R1 ‐50%
R2 100%
R 0%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 119
119 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 120
120

30
5/19/2020

How can we have a huge gain but Exercise: Calculate the exact TWRR
show a 0.00% return? for the following scenario
5,000 110,000 1 2 # Shares Price Value
R ...    1    1  0.00% 6/1 Starting position       1,000 $10 $   10,000
10,000 55,000 2 1
6/15 Price rises to $15, we decide to buy more
# Shares Price Value
6/1 Starting position       1,000 $10 $   10,000 6/15 Purchase     10,000 $15 $ 150,000
6/15 Price drops to $5, and so we buy more
6/30 Price returns to starting price
6/15 Purchase     10,000 $5 $   50,000
6/30 Price returns to starting price 6/30 Ending position     11,000 $10 $ 110,000
6/30 Ending position     11,000 $10 $ 110,000
Unrealized loss     10,000 $5 $  (50,000)
Unrealized gain     10,000 $5 $   50,000
R1 ‐50%
R2 100%
R 0%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 121
121 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 122
122

Solution How can we have a huge loss but


R
Vt 1
1 
1,000  15

110,000
1 
15,000 110,000
  1  0.00%
show a 0.00% return?
Vt 10,000 11,000  15 10,000 165,000
15,000 110,000 3 2
# Shares Price Value R ...    1    1  0.00%
6/1 Starting position       1,000 $10 $   10,000
10,000 165,000 2 3
6/15 Price rises to $15, we decide to buy more # Shares Price Value
6/1 Starting position       1,000 $10 $   10,000
6/15 Purchase     10,000 $15 $ 150,000
6/15 Price rises to $15, we decide to buy more
6/30 Price returns to starting price 6/15 Purchase     10,000 $15 $ 150,000
6/30 Ending position     11,000 $10 $ 110,000 6/30 Price returns to starting price
Unrealized loss     10,000 $5 $  (50,000) 6/30 Ending position     11,000 $10 $ 110,000
R1 50% Unrealized loss     10,000 $5 $  (50,000)
R1 50%
R2 ‐33%
R2 ‐33%
R 0%
R 0%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 123
123 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 124
124

31
5/19/2020

Our recommendation

• Since the manager usually controls internal cash flows
• And since the TWRR eliminates the effect of these 
flows, therefore not reflecting these decisions
• And since the IRR is sensitive to the timing and size of 
flows
• The IRR is the preferred method for sub‐portfolio ROR Any
Questions?
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 125
125 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 126
126

Handling Income First quarter’s returns
Why Accruals Make Sense Starting position - first quarter
Accrual Cash
100,000 100,000
Initial transaction: buy 6% bond with
2 month's accrued income
Principal amount of transaction 99,000 99,000
Accrued interest paid (2 months) 1,000 1,000
Ending position (end of 1st quarter)
Bond's market value 99,000 99,000
Accrued interest (5 months) 2,500 0
Total market value 101,500 99,000
First quarter return 1.50% -1%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 127
127 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 128
128

32
5/19/2020

2nd quarter’s returns & returns for half‐year How do we handle equity accruals?
Accrual Cash
Starting position - second quarter 101,500 99,000
Activity
At end of month 2, receive interest
(1/2 of year's interest) 3,000 3,000
Ending position (end of 2nd quarter)
Bond's market value 99,000 99,000
Cash from bond interest 3,000 3,000
Accrued interest paid (2 months) 1,000 0
Total market value 103,000 102,000
Return for 2nd quarter 1.48% 3%
Return for two quarters 3% 2%

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 129
129 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 130
130

Impact of dividend announcement on stock price 
Let’s look at “ex date’s” impact on our security
and return
This is the point Stock price drops, as it no longer
where we recognize reflects pending dividend payment
an increase in the Dividend announced Without an accrual, we’ll see a drop
security’s return in performance until dividend received

Stock price prior to


dividend announcement “Ex date”: purchasers of
security after this date
not entitled to dividend
0.00 0.00
We have an appreciation in the value Without an accrual, we’ll see a drop in
Time of the security, that reflects the Time performance (on ex date) and a rise in
pending dividend payment performance (on pay date)
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 131
131 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 132
132

33
5/19/2020

Apply an accrual, to capture and retain 
increase from pending dividend until pay date

On “ex date,” apply an accrual


dividend received; we
On “pay date,” reverse the accrual
0.00
By applying and retaining an accrual until
pay date, we avoid the return going up (at
Time dividend announcement), down (on ex date),
and up again on pay date

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 133
133 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 134
134

DIFFERENT APPROACHES TO Start‐of‐Day 


Issues

• Problem with final day (when funds are withdrawn)
• Problem with withdrawals
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 135
135 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 136
136

34
5/19/2020

End‐of‐Day  Start‐of‐Day 
Issues policy issues
• Standard Modified Dietz 
Formula:
VE  V0  C
R
V0  W  C
• Since treating flows as 
start‐of‐day, “W” is 1, the 
formula is simplified:

• Problems with initial day (when funds are contributed) VE  V0  C
RSOD 
V0  C
• Problem with contributions
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 137
137 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 138 138

Start‐of‐Day policy issues Start‐of‐Day policy issues
• Final day, when remaining assets are transferred Losses/gains from the sale of an asset and the 
• Portfolio starts the day with $10,000
• Client directs the transfer of all assets, and so we 
subsequent withdrawal of the proceeds, with 
have a withdrawal of $10,000 the overall loss (or gain) attributed to what the 
V E  V0  C 0  10,000  ( 10,000) 0 portfolio began with, less the withdrawal 
RSOD   
V0  C 10,000  ( 10,000) 0 amount
• 0/0=indeterminant

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 139
139 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 140
140

35
5/19/2020

Start‐of‐Day policy issues End‐of‐Day policy 
issues
• Portfolio holds 1,000 shares of XZ valued at $100 per share; total 
 Standard Modified Dietz 
start of day value is $100,000.  Formula:
• Half the position is sold; for $90/share VE  V0  C
• The sale’s proceeds ($45,000) are withdrawn and treated as a  R
V0  W  C
start‐of‐day event.
 Since treating flows as 
• The remaining shares are valued at the same $90 price at the end  end‐of‐day, “W” is 0, the 
of the day. formula is simplified:
45,000  100,000  ( 45,000)  10,000 V  V0  C
RORSOD    1818%
. R EOD  E
100,000  ( 45,000) 55,000 V0

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 141
141 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 142
142

End‐of‐Day policy issues And speaking of being undefined
• Initial Day of investing VE  V0  C
REOD 
• VB =0 V0
• C=100,000 101,000  0  100,000

• VE = 101,000  0
1,000
  Undefined
0

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 143
143 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 144
144

36
5/19/2020

End‐of‐Day policy issues The following approach eliminates 
these problems:
• Gains/losses from contributions which are attributed 
solely to what the portfolio began with. E.g., • Treat all inflows as start‐of‐day events
• Starting value = $100,000 • Treat all outflows as end‐of‐day events
• Cash flow = $100,000
• Gains on both during the day = $1,000, meaning total gain = 
$2,000 V  V  C 202,000  100,000  100,000 2,000
REOD  E 0
   2.00%
V0 100,000 100,000

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 145
145 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 146
146

What about weighting each flow during the day???
 E.g., mid‐day (W=0.5)
 Or, more detailed:
 W=0.0, 0.1, 0.2,
… 0.9, 1.0

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 147
147 Copyright © The Spaulding Group, Inc. 2020 148

37
5/19/2020

Benchmarks: 
Common in every day life
“Factoid,” as seen on a movie screen
“Anthony Hopkins recorded
his narration of 
How The Grinch Stole Christmas
in a single day”
My question: Was that good or bad?

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 149
149 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 150
150

Peter L. Bernstein, Author & Legendary Investor


From the front page of the  “The primary purpose of a
5/11/07 WSJ: benchmark is to set a realistic,
attainable performance
• “A New Orleans report  standard, so that, by closely
says two thirds of high  replicating a manager’s style
and chosen risk level, any
school seniors failed  short- or long-run differences in
graduation exams since the  performance that arise can be
state’s schools takeover  attributed to the manager’s
following Katrina.” active decisions, and thus –
eventually –
• Our conclusion? to manager skill”
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 151
151 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 152
152

38
5/19/2020

Properties of benchmarks
• Unambiguous
• Absolute • Investable
• Market Indexes • Measurable
• Peer Groups • Appropriate
• Custom (we’ll use Performance Opportunity  • Reflective of current investment opinions
Distributions “PODs”) • Specified in advance
• Owned
Source: CIPM materials

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 153
153 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 154
154

Unambiguous Investable
The identities and weights of securities It is possible to forgo active management and
constituting the benchmark are clearly simply hold the benchmark
defined.

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 155
155 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 156
156

39
5/19/2020

Measurable Appropriate
The benchmark's return is readily calculable The benchmark is consistent with the
on a reasonably frequent basis manager's investment style or area of
expertise

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 157
157 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 158
158

Reflective of current Specified in advance


investment options The benchmark is specified prior to the start
of an evaluation period and known to all
The benchmark is consistent with the
interested parties
manager's investment style or area of
expertise

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 159
159 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 160
160

40
5/19/2020

Owned Absolute Benchmarks
The investment manager should be aware of
and accept accountability for the
constituents and performance of the
benchmark. It is encouraged that the
benchmark be embedded in and integral to
the investment process and procedures of
the investment manager.

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 161
161 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 162
162

Will Rogers, on investing Absolute
“Investing is easy,…, just buy • Usually set by the client
• The income needed by a 75‐year old widow
some good stock and hold it • The income needed for a pension fund, as defined by the 
‘til it goes up, then sell it.” plan’s actuaries
“What do you do if it don’t go • Also known as:
up?” • Minimum funding requirement (MFR)
“Then, I don’t buy it.” • Minimal acceptable return (MAR)
• Liability‐related return
Will Rogers (November 4, 1879 – August 15, 1935)
American cowboy, comedian, humorist, social
• Appropriate for “goal‐oriented” investing
commentator, vaudeville performer and actor.

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 163
163 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 164
164

41
5/19/2020

How do absolute  • Unambiguous (yes)
benchmarks measure up to  • Investable (no)
our criteria? • Measurable (yes) 
• Unambiguous  • Appropriate (yes)
• Investable • Reflective of current investment opinions (no)
• Measurable • Specified in advance (yes)
• Appropriate • Owned (no)
• Reflective of current investment 
opinions
• Specified in advance
• Owned
Source: CIPM Prep Materials
Source: CIPM Prep MaterialsCopyright © The Spaulding Group, Inc. 2020 165 Copyright©©The
TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 166
Copyright © The Spaulding Group, Inc. 2020 165 Copyright 166

Indexes as benchmarks Role of the Index 
• Provide an indication of how the “market” did
• Serve as a benchmark for comparison

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 167
167 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 168
168

42
5/19/2020

Defining an Index What do index providers do?
Index make‐up / composition • Differentiate themselves based upon their 
• broad / narrow market construction rules
• across/within country
• across/within region
• They create intellectual property
• across/within industry
• security type(s)
• risk level(s)
• maturity level(s)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 169
169 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 170
170

Index calculation outputs Weighting methods
‐ not always published
• Price: Securities with higher prices get heavier 
• Price & total return indices
weightings
• Index yield & capitalization • Equal: All stocks/countries get the same weight
• Weight of sub‐index in main index • GDP: Weights countries by their economic output
• Weight of stock in index • Market Capitalization: Weight by size of company
• Stock return (w/ or w/o income) • Float: Weight securities by what is actually 
available on the open market
And, the rules may change!
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 171
171 Copyright © The Spaulding Group, Inc. 2020 172

43
5/19/2020

Methodology Differences
• Stock selection
• Total return calculation
• Free float
• Industry classification
• Size bands & Style indices
• Rebalancing frequency • Market indexes do not reflect transaction costs
• At times, their makeup may not be known

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 173
173 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 174
174

EVALUATING INDEXES
• Objectivity and availability of construction rules
• History: index’s inception date
• Market Coverage: percent of total stocks in universe included; 
percent of market cap included
• Concentration: accurately measure market? Excessively 
influenced by a handful of companies?
• Turnover
• Weighting Method: Price, Cap , Equal, GDP,Float
• Cost!
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 175
175 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 176
176

44
5/19/2020

S&P 500: A bit concentrated, yes? Market coverage:
Diminishing Value of Adding Stocks

100%

Cumulative Percent of Index


500 = 83% 1,000 =
90% 95%
80%
70%
60%
50%
40%
30%
20%
10%
0%

1 50
3 00
4 50
6 00
7 50
9 00
1

1 ,0
1 ,2
1 ,3
1 ,5
1 ,6
1 ,8
1 ,9
2 ,1
50
00
50
00
50
00
50
00
Data Source:MSCI, FactSet
MSCI ACWI Free as of December 31, 2002
Number of Stocks Included

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 177
177 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 178
178

The impact of turnover Float weighting


• Transaction costs can vary considerably, from market to  • Float is the amount of shares available on the open market
market • With float weighting, each company is weighted according 
• Emerging markets and small cap, for example, may have  to how much of its stock is available to be sold to the 
higher costs to acquire assets than large cap or more  public.
developed markets • Advantages
• Consequently, turnover differences, from one index to  • Large portion of some big companies unavailable
another, might translate into more or less transaction costs,  • Fairer reflection of what is actually available
to respond to changes in the index’s makeup • Makes a large difference for some industries like 
• Something to consider, when comparing competing indexes telecommunications

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 179
179 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 180
180

45
5/19/2020

Free float characteristics Shares held by insiders


Strategic Shareholders • Example of shares held by insiders
• Government holdings a/o March 2004, 38.9% of Wal‐Mart and 48.1% of Coca‐Cola held 
by insiders
• Holdings by other companies (cross holdings)
• With float weighting, these shares wouldn’t be included in 
• Holdings by principle officers / board the index weights.
• Holdings by individuals / families of founders
• Holdings by employees (retirement savings)
Source: “Money managers praise S&P for float‐adjusted index weightings,” Pensions & Investments, March 8, 2004

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 181
181 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 182
182

Who determines which index to


Cost! use?
Evidence suggests it’s the client, in the
case of institutional management (though
their consultant probably has input, too!

Many market index


prices have risen!
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 183
183 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 184
184

46
5/19/2020

There’s a lot of ignorance about index Who pays for the increased costs?
costs and restrictions! Typically, two parties:
- Some managers are unaware that they ‐ The portfolio manager
have to pay ‐ The custodian
- Many (most?) clients are not aware of how Many custodians are now passing some of 
expensive they can be these costs along to their clients, especially 
- Some clients and managers think they can when custom indexes are requested
derive their own custom indexes … check
your license!
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 185
185 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 186
186

Blending of Indexes
Does it really matter which index you
use for a specific market? • A mix of two or more indexes
• Common way to represent balanced portfolios
Research* The Spaulding Group did with three • A way to represent style across multiple indexes
custodians (BNY Mellon, Northern Trust, & State • Critical to ensure that your license agreements allow 
Street) suggest that it really doesn’t matter. blending
• Decisions need to be made regarding the rebalancing 
* See Barney, Johnson, et al. 20015/2016. “Are All Market Indexes Created Equal?” The Journal of Performance timing
Measurement. Winter.

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 187
187 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 188
188

47
5/19/2020

Which percent (%) to use? What should the blended


• Base off of the actual allocation or target allocation? benchmark’s rebalancing timing be?
• This is essentially a • It should align with the rebalancing of the associated 
• Tactical vs. strategic allocation decision portfolio(s)/composite
• We believe it should be based off the strategy. • If the portfolio(s)/composite rebalances monthly, 
rebalance monthly

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 189
189 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 190
190

A Simple Blending Example WHEN IT COMES TO


INDEXES …
• Equity Index = S&P500 (60% Target); Return = 3.00%
• Fixed Income Index = 
Bloomberg Barclay’s Agg (40% Target); Return = 5.00%
n
BB   A i r i  60%  5.00%  40%  3.00%  4.20%
i 1

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 191
191 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 192
192

48
5/19/2020

Choosing your indexes Concluding remarks about indexes …


• The index decision is important
• Attribution highlights differences between
• Indexes that seem similar may be very different
the portfolio and benchmark. when examined closely
• Seemingly similar indexes can have very • Faulty indexes lead to bad investment decisions
different returns & soured client relations
• The index is the basis for the entire • Using an inappropriate index may lead to
analysis regulatory scrutiny
• If the index is wrong, the attribution has • GIPS verifiers strive to ensure the index is
little value appropriate for the composite
• Cost is becoming an important issue to consider
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 193
193 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 194
194

• Unambiguous (yes)
How do market indexes  • Investable (yes)
measure up to our criteria? • Measurable (yes) 
• Unambiguous  • Appropriate (no)
• Investable • Reflective of current investment opinions (yes)
• Measurable • Specified in advance (yes)
• Appropriate • Owned (yes)
• Reflective of current investment 
opinions
• Specified in advance
Source: CIPM Prep Materials • Owned
Source: CIPM Prep Materials
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 195
195 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 196
196

49
5/19/2020

Peer Groups
• Compares manager with others
• Lipper and Morningstar universes
Peer Groups  • Consultant groupings
(a.k.a. universes) 
as benchmarks

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 197
197 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 198
198

Peer group example Advantages of Peer Universes
• Measurement is against active managers, not an 
unmanaged index
• Resolves managers’ reservations about benchmark not 
being a real portfolio

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 199
199 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 200
200

50
5/19/2020

Disadvantages of Peer Universes Questionable makeup...
• Questionable make‐up of the universe The Kramer effect
• Not an objective benchmark
• Holdings are not available
Cannot measure active “bets” vs. universe
• Cannot identify median manager ahead of time
• Different median manager for each time period
• Unknown ROR calculation method
• Unavoidable “Survivorship bias”
https://www.youtube.com/watch?v=ad8M_QrSf40

Copyright
Copyright © The Spaulding © The
Group, Inc.Spaulding
2020 Group, Inc. 2020 201
201 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 202
202

Also, Survivorship Bias
To what is the manager expected to add value? Without a 
valid reference point, superior performance remains an  Manager 8

elusive notion. Placing above the median of a universe of  Manager 7

investment managers or funds may be a reasonable  Manager 6

investment objective, but the performance of a particular  Manager 5

manager or fund is not a suitable performance  Manager 4

benchmark that can be used to assess investment skill. Manager 3

Manager 2

Manager 1

Jan-93 Jun-94 Oct-95 Mar-97 Jul-98 Dec-99

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc. CIPM
2020 study materials
Inc.2020 203
203 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 204
204

51
5/19/2020

• Unambiguous (no)
How do peer groups  • Investable (no)
measure up to our criteria? • Measurable (yes) 
• Unambiguous  • Appropriate (no)
• Investable • Reflective of current investment opinions (no)
• Measurable • Specified in advance (no)
• Appropriate • Owned (no)
• Reflective of current 
investment opinions
• Specified in advance
Source: CIPM Prep Materials • Owned
Source: CIPM Prep Materials
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 205
205 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 206
206

Custom – e.g., Performance  Custom benchmarks: screen out securities 


Opportunity Distributions (PODS) the manager wouldn’t buy; include only 
• An example of a “custom” benchmark those that meet the manager’s criteria
• Proponents suggest that PODS eliminate the 
shortcomings of indexes and benchmarks Universe of securities

• Based on manager’s definition of his/her market
Criteria / Rules

• E.g., I invest only in auto stocks

Opportunity Set

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 207
207 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 208
208

52
5/19/2020

The universe of  Manager Defines His/Her Rules


investment opportunities
• Target Securities
• Auto stocks
• Typical allocation
• 5% positions; 100% invested; ‐> 20 securities

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 209
209 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 210
210

Manager’s Portion of Universe Speaking of Venn Diagrams … (from 8/9/18)
(Performance Opportunity)

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 211
211 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 212
212

53
5/19/2020

Create Portfolios Plot the returns; compare the 
Measure Returns portfolio’s with the average
• Identify all the combinations of opportunities from 
manager’s segment of the universe
• Measure returns for these and the manager’s selection
• Calculate the average
• Compare 
Average

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 213
213 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 214
214

PODS ‐ Summary How do custom benchmarks 


measure up to our criteria?
• Seems to have some merit / value • Unambiguous 
• Takes away some of the shortcomings of other  • Investable
methods
• Measurable
• Requires heavy‐duty processing capabilities • Appropriate
• May be good to supplement indexes / peer  • Reflective of current 
groups investment opinions
• Hasn’t caught on • Specified in advance
• Owned
Source: CIPM Prep Materials

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 215
215 Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 216
216

54
5/19/2020

• Unambiguous (yes) Summary: How do the four most 
• Investable (yes)
• Measurable (yes) 
common benchmarks rank?
• Appropriate (yes) Benchmarks
Market Peer
• Reflective of current investment opinions (yes) Criteria: Absolute Indexes Groups Custom
• Specified in advance (yes) 1 Unambiguous Yes Yes No Yes
2 Investable No Yes No Yes
• Owned (yes)
3 Measurable Yes Yes Yes Yes
4 Appropriate Yes No No Yes
5 Reflective of opinions No Yes No Yes
6 Specified in advance Yes Yes No Yes
Source: CIPM Prep Materials
7 Owned No Yes No Yes
Copyright © The Spaulding Group, Inc. 2020 218
Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 217
217 Copyright © The Spaulding Group, Inc. 2020 218

What we’ve covered
Questions?
• What benchmarks are
• Absolute
• Indexes (including a way to blend quarterly)
• Peer Groups
• PODs – Performance Opportunity Distributions
• Pros / Cons of each

Copyright©©The
Copyright TheSpaulding
SpauldingGroup,
Group,Inc.
Inc.2020
2020 219
219 Copyright © The Spaulding Group, Inc. 2020 220

55
5/19/2020

John Simpson
will cover …

Performance 
Attribution!

Copyright © The Spaulding Group, Inc. 2020


Copyright © The Spaulding Group, Inc. 2020 221
221 Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 222
222

FOR JOINING David D. Spaulding,


DPS, CIPM
US TODAY DSpaulding@SpauldingGrp.com
We hope you found @DSpauldingAtTSG
www.SpauldingGrp.com
the session of interest www.linkedin.com/in/daviddspaulding

Blog: http://www.spauldinggrp.com/investment-performance-guy/
Copyright © The Spaulding Group, Inc. 2020
Copyright © The Spaulding Group, Inc. 2020 223
223 Newsletter: https://wc111.infusionsoft.com/app/form/na15
© The Spaulding Group. 2016
Copyright © The Spaulding Group, Inc. 2020 224

56
5/19/2020

The Spaulding Group, Inc. is the leading provider of


performance and risk measurement products and services
to the investment industry. It is the fastest growing GIPS
verification firm, hosts the annual PMAR conferences and
Performance Measurement Forum and Asset Owners’
Roundtable, provides consulting services (which include
operations reviews and software certifications), publishes
The Journal of Performance Measurement® and the
Spaulding Series of books, and conducts training classes,
both open enrollment and in-house, which can be
customized to meet our client’s needs.

Copyright © The Spaulding Group, Inc. 2020 225

57

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy