Equity Compensation Issue Brief

May 2016

Tax Settlement
Forecasting: The Next
Big Thing in Stock-Based
Compensation
Accounting
Introduction
With Accounting Standards Update No. 2016-09 (ASU 2016-09), FASB intended to
simplify share-based accounting. But at least in one area, brace for more complexity.
The APIC pool will no longer buffer excess tax benefits and deficiencies from the income
statement. Instead, excess benefits and deficiencies will flow directly through the
income statement, giving rise to volatility in net income and the effective tax rate.

The solution? No different from how you deal with any other source of uncertainty:
Forecast it, then flex the forecast across multiple scenarios. That way, you’ll understand
and socialize how sensitive the final result is to changes in the relevant input
assumptions.

Already, many companies are implementing tax settlement forecasting for insight into
future excess benefits and deficiencies under various stock price and other scenarios.

To help you prepare, here’s an introduction to tax settlement forecasting. We’ll take you
through the four basic steps of the tax settlement forecasting process, followed by
examples of what a final forecast can look like. But first, let’s review the current
deferred tax accounting model in ASC 718 and what it means to eliminate the APIC pool.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 1
The Current Deferred Tax Accounting
Model
The basic mechanics of tax accounting for equity awards entail setting up a deferred tax
asset (DTA). You base the DTA on the cumulative book expense for awards that are
expected to result in future tax deductions, and reverse that DTA when settling the
award for tax purposes. Differences between these respective measures of value (taken
at different times) are typically recorded in additional paid-in-capital (APIC). There are
also a host of specific rules and issues on top of the basic mechanics.

With the release of ASC 718 in 2004, companies were required to recognize in their
financial statements compensation cost arising from the issuance of share-based
compensation instruments under a fair value methodology. In contrast, the Internal
Revenue Service (IRS) tax code has always treated share-based payment awards as an
expense.

ASC 718 requires measuring compensation cost using fair value principles and
recognizing this cost over the requisite service period (generally the vesting period). But
the tax deduction is recorded on a company’s tax return only when a settlement event
occurs (a “taxable event”). The tax benefit that the company receives at settlement and
recognizes in its tax return is generally equal to the intrinsic value of the award on the
date of the settlement event.

Most equity awards give rise to a DTA because expense under GAAP is calculated and
allocated differently from under tax rules. A key principle of ASC 740 is that the tax
benefit of a deduction that is available for both book and tax purposes, but differs in
terms of timing (a so-called temporary difference), should be recognized on the books at
the same time as the book expense.

The way to do this is to record a DTA on the balance sheet. The DTA represents a future
deductible amount that will produce tax savings at a later date. At the same time you
record the DTA, you also record a deferred tax benefit in the income statement. That
reduces current period tax expense. At the close of each reporting period, the balance in
the DTA should equal the cumulative recorded compensation cost for outstanding
awards multiplied by the relevant corporate tax rate.

Once you settle an award and realize the actual tax benefit (if there is one), you reverse
the DTA from the balance sheet. If the actual tax benefit exceeds the DTA, you have an
excess benefit (aka “tax windfall”). Otherwise, you have a tax deficiency (“tax shortfall”).
Prior to ASU 2016-09, excess benefits built up the APIC pool and tax deficiencies

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 2
 depleted it. The APIC pool is an off-balance sheet pool of cumulative excess benefits
that shields the income statement from tax deficiencies. Only when the APIC pool is
depleted down to zero will tax deficiencies flow through to tax expense in the income
statement.

A World Without the APIC Pool

Settlements almost always occur at amounts greater than or less than the initial fair
value. The APIC pool’s role of buffering the P&L from windfalls and shortfalls has been
generally appreciated in our present-day world where even favorable budget-to-actual
variances are frowned upon.

As the APIC pool prepares to breathe its last, volatility in earnings and the effective tax
rate are around the corner. The effect on the income statement is positive during
periods of rising stock prices and negative during periods of falling stock prices.

Getting Ready for ASU 2016-09

Companies are preparing for the tax-specific components of ASU 2016-09 in two ways.
First, they’re streamlining and automating their tax reporting.

Corporate tax departments are busy and have many competing priorities. For some
companies, this has created inefficiencies in deferred tax tracking. In our survey of
stock-based compensation accounting best practices, we saw considerable disparity in
how companies split roles and responsibilities between corporate accounting and tax.
Handoff risks are greatest when the process lacks tight integration between the two
departments. In the past, this meant potential misstatements to APIC. Now the
misstatement risk is to the P&L.

The second way companies are getting ready is to set up a tax settlement forecasting
process. If excess benefits and deficiencies are to flow directly through the income
statement—and if random events like the stock price and the timing of option exercises
are going to trigger them—then companies need a way to forecast and understand the
potential effects. If EPS could change by $0.03 next quarter or next year, you need to
understand that now, along with what could cause it to happen.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 3
Tax Settlement Forecasting: A Step-by-
Step Guide
Tax settlement forecasting has four basic steps:

1. Project future tax settlement events over time

2. Layer in hypothetical future grants
3. Compute excess benefits or deficiencies at settlement
4. Run scenarios to factor uncertainty into your forecast

Let’s look at each step in more detail.

1. Project Future Tax Settlement Events Over Time

Restricted stock award and unit releases are straightforward to project because the
dates are known. But the timing of stock option exercises and possibility of expiration is
not known. That creates complexity for heavy option granters.

Assumptions need to be made about the exercise frequency and gain, as well as the
possibility of expiration out-of-the-money.

Depending on the materiality of your option program, you may opt for a simple method
or an advanced method to develop an exercise rate assumption. A simple approach
usually involves a historical analysis of the average rate of exercise each year. At the
other end of the spectrum, you can perform predictive modeling to develop a custom
rate for each distinct grant based on its current moneyness and remaining life.

While projecting typical tax settlement events, remember to take potential forfeitures
into consideration. For instance, suppose you have 100 shares for 10 participants (each
person holds 10 shares). If you expect a single participant to terminate before the
awards vest, then 10 shares will forfeit and have no taxable event at all. In this case, the
maximum shares to be settled are 90 instead of 100. You can predict forfeitures via a
standard dynamic forfeiture rate application. (This is another reason we have a slight
but general preference for companies to continuing applying a forfeiture rate, at least
for forecasting.)

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 4
 2. Layer in Hypothetical Future Grants
Creating hypothetical future grants is complicated enough to warrant its own Issue
Brief. You can take last year’s grant and grow it by some percent (a top-down approach).
You can also build the assumption by modeling hires, terminations, and promotions (a
bottom-up approach). There are any number of variations on these techniques.

For any forecast that runs out past the end of the current year, hypothetical grants are
essential. Simple top-down approaches are adequate when the stock compensation
program is relatively stable year over year. Growth companies that are actively
launching and divesting business units, and have frequent employee mobility, need a
more granular bottom-up method. Multinational companies should at least project
grants at the country level so they can apply country-level tax rates.

Of course, any work here in developing a framework for modeling hypothetical grants
carries over neatly to your process for forecasting expense and EPS.

3. Compute Excess Benefits or Deficiencies at Settlement

The calculation for the excess benefit itself is simple. It’s the difference between the
potential tax benefit and the DTA reversal at settlement. To break it down:

Excess Benefit/Deficiency = (Intrinsic Value at Settlement - GAAP Fair Value) * Shares

Settled * Tax Rate

The hard part is deciding what your base case will be. As we’ll explain in the fourth step,
numerous variables drive the final result. They include:

 Stock price
 Exercise rate for options
 Tax rate expectations
 Forfeiture rate
 Performance outcome expectations (for both performance and market
conditions)

Nobody can predict the stock price. But you can count on your colleagues in Treasury or
Financial Planning and Analysis (FP&A) for the company’s best-guess estimate, since it’s
a common assumption in many other forecasts your company performs. Your tax team
may have expectations of future tax rates, or they may ask you to simply assume
continuity in the rate. A study will need to be performed to develop an exercise rate and
forfeiture rate assumption.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 5
 Finally, most companies take the multiplier they apply for EPS contingently issuable
share purposes and use it to model the expected outcome on their performance-based
awards—at least for the base case assumption.

When done, your base case will show how removal of the APIC pool will affect the
income statement in the future. You’ll want to set this expectation among your
stakeholders.

4. Run Scenarios to Factor Uncertainty Into Your Forecast

Bad news, though. Your base case will be wrong.

That’s why scenario analysis is so important. The sheer quantity and materiality of
variables beyond your control will cause meaningful budget-to-actual variances.
Scenario analysis helps you communicate just how many potential variances there are,
and how sensitive the final outcome is to the variables and assumptions driving the
calculation.

For example, it’s particularly interesting to use scenario analysis to determine boundary
cases, such as how much the stock price needs to drop in order to produce a net tax
deficiency. If it takes only a 6% drop, that’s very different from a 26% drop.

Of the five key variables—stock price, exercise rate, forfeiture rate, tax rate, and
expected performance outcomes—stock price is usually the most sensitive. It drives the
intrinsic value, and thus most directly affects the direction and size of windfalls or
shortfalls.

Try to obtain from Treasury or FP&A not only a base case stock price assumption, but
also a favorable and unfavorable scenario. If they cannot provide one, apply a plus and
minus 10% factor to the base case. The process may be iterative as you try different
stock price premiums and discounts in an effort to disentangle the sensitivities.

Layer Thoughtful Precision into Your Forecasts

After flexing the stock price, you can get more sophisticated or stop there. If you begin
flexing other variables, you’ll need to think about whether it makes sense to model all
the combinations. For example, if you have three stock price assumptions and three
performance outcome scenarios, in total you could model nine possible outcomes.

A more practical view is to think about the correlations between assumptions. For
example, a heavy option granter may wish to connect a favorable stock price
assumption to a higher exercise rate assumption. Similarly, for TSR-based awards, a

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 6
 favorable stock price scenario could be linked with a higher performance outcome
expectation.

Think Carefully About Market and Performance Awards

Be careful with market and performance awards. Multipliers, if applicable, should be

tied to the expected payout on the settlement date for the specific scenario being
considered. These may be different from the current multipliers used for expense
(current expected payout) and diluted EPS purposes (payout if the awards were settled
today).

Make Sure the Scenarios are Interesting

It’s possible to run three scenarios and find that they all result in the same impact to
EPS. This is interesting insofar as it shows that your earnings aren’t very sensitive to your
assumptions. It may also be interesting to find what it takes to move EPS by a
meaningful amount. Or, for example, how the most optimistic stock price and
performance outcome, plus the most aggressive exercise rate, affect EPS. In our
experience, senior management appreciates this level of scenario bending and
stretching.

None of this is to endorse complexity for the sake of complexity. Forecasting at current
estimates using base case assumptions can establish a reasonable set of status quo
numbers. You might find that this is enough for short-term needs such as quarterly
forecasting. Then, once a year, you could run a deeper analysis to see which variables
matter the most and whether the sensitivities are changing.

Forecasting is management accounting, so the only rule is to do what works and delivers
insight within the organization.

Extend Your Forecasting to Cash Burn Analysis

For companies that actively forecast and track cash inflow and outflow from option
exercises and tax withholding payments, there may be another new piece to the puzzle.
The cash from the exercise of options is a material cash inflow. But withholding and
paying the tax under net settlement schemes can be a material cash outflow.

That’s not new, but this is: If you begin withholding at higher levels (given how ASU
2016-09 relaxes the liability trigger for tax withholding), you might be writing much
larger checks to the IRS. You probably don’t need to do this type of analysis quarterly,
but you should at least consider doing it periodically.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 7
 Other Considerations

Finally, there’s the marriage between forecasting expense and forecasting tax
settlement. Finance and tax departments are also asked to help with Schedule M-1. This
reconciliation process requires an explanation of the difference in book expenses and
tax expenses. Having an accurate measure of both will limit noise when the actual M-1
is completed.

Excess Tax Benefit in Forecast EPS—Is That the Same

Thing?
If you already have a working forecast EPS process in place, you may notice that there is
an excess tax benefit (ETB) as one of the three components of total proceeds in the
treasury stock method. It’s important to keep in mind that the ETB will no longer be part
of the total proceeds post adoption of ASU 2016-09.

Further, the underlying treasury stock method is based on an “as-if” assumption that
calculates the maximum dilution impact if all unsettled awards are settled in the current
period at the average stock price. In contrast, the true tax benefit is based on
transactions for awards legally vested or exercised by participants.

Finally, the as-if settlement in EPS is done on a weighted-average basis. For instance,
assume 100 shares are settled on June 30, 2017. The ETB for year-to-date EPS purposes
would be based on 50 shares since the award was outstanding for the first half of the
year only. Once the true tax benefit occurs, it will be for 100 shares.

In short, the underlying process or architecture used to perform tax settlement

forecasting and EPS forecasting is similar, but the results and inner assumptions are
quite different.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 8
What a Tax Settlement Forecast Can
Look Like
Since tax settlement forecasting is part of management accounting, the optimal format
and work product will differ by organization. Again, the one rule is to deliver insight.

A basic template might show the P&L impact of settlements with a waterfall of
outstanding awards by grant year and expected settlement date. You might also slice it
by award type or other relevant indicative data.

Table 1 illustrates a case of increasing stock prices. In this example, stock options were
phased out in 2010. For the sake of simplicity, time-based and performance-based RSUs
aren’t shown separately—but normally they should be. By itself, the table is a bit of a
head-scratcher because it’s unclear why the numbers are what they are. That’s why the
process we just stepped through is so important.

Forecast Year 2016 2017 2018 2019 2020

ABC Co.
Assumed Stock Price $35 $42 $48 $52 $55
EXPECTED P&L CREDIT/(DEBIT) BY GRANT YEAR & AWARD TYPE
2006 Grants ($963)
2007 Grants ($744) ($499)
Stock
2008 Grants ($88) $158 $368
Options
2009 Grants $131 $376 $586 $726
2010 Grants $350 $595 $805 $945 $1,050
2013 Grants ($525)
2014 Grants ($175) $70
RSUs 2015 Grants $175 $420 $630
2016 Hypo Grants $245 $455 $595
2017 Hypo Grants $210 $350 $455
Totals ($1,838) $1,365 $3,054 $2,616 $1,505
Table 1: Sample waterfall summary of P&L impact of expected windfalls and shortfalls under a scenario of increasing
stock prices

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 9
 Next, we show the inverse case in Table 2. Not surprisingly, the numbers worsen.

Forecast Year 2016 2017 2018 2019 2020

ABC Co.
Assumed Stock Price $55 $52 $48 $42 $35
EXPECTED P&L CREDIT/(DEBIT) BY GRANT YEAR & AWARD TYPE
2006 Grants ($263)
2007 Grants ($44) ($149)
Stock
2008 Grants $613 $508 $368
Options
2009 Grants $831 $726 $586 $376
2010 Grants $1,050 $945 $805 $595 $350
2013 Grants $175
2014 Grants $525 $420
RSUs 2015 Grants $875 $770 $630
2016 Hypo Grants $595 $455 $245
2017 Hypo Grants $210 $0 ($245)
Totals $3,763 $3,815 $3,054 $1,216 $105
Table 2: Sample waterfall summary of P&L impact of expected windfalls and shortfalls under a scenario of decreasing
stock prices

Table 3 is another example of the P&L impact of expected settlements under different
scenarios. It shows the correlation among the changing stock prices, exercise rates, and
performance payout movements. In this case, expense is very sensitive to the
combination of factors. For simplicity, the columns all pertain to a single year (e.g., the
next fiscal year) and capture the spectrum of favorable and unfavorable assumptions
applied.

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 10
 Unfavorable Moderate Favorable
Assumed Stock Price $35 $42 $48 $52 $55
ABC Co.
Payout for Market Awards 50% 75% 100% 150% 200%

2006 Grants—$50 Strike Price 0% 0% 0% 5% 5%

2007 Grants—$45 Strike Price 0% 0% 5% 10% 10%
Option
2008 Grants—$30 Strike Price 5% 20% 30% 40% 50%
Exercise Rate
2009 Grants—$25 Strike Price 20% 30% 40% 50% 60%
2010 Grants—$20 Strike Price 30% 40% 50% 60% 70%
EXPECTED P&L CREDIT/(DEBIT) FOR 2017 BY AWARD TYPE
2006 Grants $0 $0 $0 ($53) ($36)
2007 Grants $0 $0 ($41) ($40) ($11)
Stock Options 2008 Grants ($13) $86 $252 $406 $525
2009 Grants $75 $258 $469 $623 $713
2010 Grants $300 $476 $690 $810 $840
2013 Grants ($525) ($280) ($70) $70 $175
2014 Grants ($175) $70 $280 $420 $525
RSUs 2015 Grants $175 $420 $630 $770 $875
2016 Hypo Grants $0 $245 $455 $595 $700
2017 Hypo Grants ($245) $0 $210 $350 $455
Market 2015 Grants ($508) ($18) $560 $1,610 $2,730
Awards
(currently 2016 Hypo Grants ($718) ($228) $350 $1,400 $2,520
expensing at
target) 2017 Hypo Grants ($963) ($473) $105 $1,155 $2,275
Totals ($2,595) $557 $3,890 $8,116 $12,286
Table 3: Sample summary of P&L impact of expected windfalls and shortfalls under five scenarios

© 2016 Equity Methods. All rights reserved. Tax Settlement Forecasting: The Next Big Thing 11
Closing Thoughts
Although FASB intended ASU 2016-09 to simplify share-based payment accounting, it
eliminated the APIC pool entirely on technical grounds. In FASB’s November 2015
meeting, Tom Linsmeier summed it up best: “The equity approach may be more
supported in 123(R), but the income approach is more consistent with how we think
about taxes overall.”

Like it or hate it, the best way to manage it is through tax settlement forecasting. A solid
process will reveal the scope of potential future income statement volatility along with
the variables that influence it.

As accounting draws closer to fair value and the income statement becomes more fluid,
analytical modeling is the best line of defense. No two cases of tax settlement
forecasting are alike. But they do have a common approach:

 Run the forecast past the current fiscal year

 Base the forecast on scenarios
 Model the scenarios to understand how much variables must move to materially
affect the final result
 Embed the tax settlement forecast into your quarterly forecasting process

For more information, please see our Issue Brief on the amendments to ASC 718, or
reach out to us directly. We welcome your questions and comments.

Daniel Hunninghake, CPA

480.428.3328
daniel.hunninghake@equitymethods.com

Takis Makridis
480.428.1203
takis.makridis@equitymethods.com

Fori Wang, CPA

480.428.3297
fori.wang@equitymethods.com

www.equitymethods.com

Equity Methods • 15300 N. 90th Street, Suite 400 • Scottsdale, AZ 85260 • 480.428.3344 • info@equitymethods.com
© 2016 Equity Methods. All rights reserved.