DISCRETIONARY EXPENDITURES:
REFINEMENTS AND CASES
Professor Brian Bushee
WHARTON ONLINE
Refinements to Discretionary Expenditure Models
• Discretionary expenditure models are very imprecise
– Still have the problem that economics of firm may have shifted to warrant a
legitimate change in expenditures
• Refinement #1: Use Distance measure to identify earnings management
motives during the year
– Company would miss earnings target without a cut in expenditures
• “Meet or Beat”: Incentive to cut expenditure to meet or beat earnings target
– Company would miss earnings target even with a cut in expenditures
• “Big Bath”: Incentive to increase expenditure to make a bad year even worse, but save
future expenses
– Company would make earnings target even with an increase in expenditures
• “Smoothing”: Incentive to increase expenditure to make a “great” year just a “good”
year, and save future expenses
• Refinement #2: Use quarterly expenses to identify unusual changes late in
the year
WHARTON ONLINE
Refinement #1: Distance Measure
• Distance Measure (for R&D)
– Distance (R&D) = (Current Pre-R&D Earnings – Prior Pre-R&D Earnings) / Prior
R&D Expense
• Pre-R&D Earnings = Pre-tax Income + R&D Expense
• Can also compute for SG&A and Advertising
2014 2015
Pre-tax Income 500 505
R&D Expense 90 84
Pre-R&D Earnings 590 589
• Distance = (589 – 590) / 90 = -0.011
– Company must cut R&D by 1.1% (from 90 to 89) to equal last year’s earnings
WHARTON ONLINE
Refinement #1: Distance Measure
• “Meet or Beat”
– If Distance is between 0 and -0.20 (or so), then company could cut enough R&D
to meet earnings target
• “Big Bath”
– If Distance is less than -0.20 (or so), then company could not feasibly cut R&D
enough to meet earnings target
• “Smoothing”
– If Distance is greater than 0, then company can increase R&D and still meet
earnings target
WHARTON ONLINE
Refinement #2: Quarterly Changes
• Some firms report quarterly amounts of expense
– In 2015, 95% reported quarterly SG&A, 30% reported quarterly R&D, 0% reported
quarterly advertising
– Year over Year (YoY) Change = (Expense this quarter – Expense same quarter
last year) / Expense same quarter last year
• If discretionary expenditures are used to manipulate annual earnings, we
would expect YoY Change in 4th quarter is different than other 3 quarters
– “Meet or Beat”: Cut R&D in Q4
– “Big Bath” or “Smoothing”: Increase R&D in Q4
• Look at pattern of quarterly changes in conjunction with Discretionary
measures and Distance
WHARTON ONLINE
Company #1: Possible “Meet or Beat”
• Dog-Techno Corp
Chg Pre-tax
Discretionary Distance YoY R&D YoY R&D YoY R&D YoY R&D income/
Year R&D (R&D) Q1 Q2 Q3 Q4 Revenue
2009 -0.003 0.083 3.4% 7.0% 6.3% 6.4% 0.002
2010 0.011 0.112 4.1% 9.3% 8.9% 4.2% 0.004
2011 -0.010 0.374 7.6% 3.3% -0.4% 3.4% 0.032
2012 -0.019 -0.033 0.7% 3.5% 13.1% 12.7% -0.010
2013 -0.054 -0.007 11.8% 8.3% 1.4% -1.4% 0.005
WHARTON ONLINE
Company #2: Possible “Big Bath”
• Rexido Corp
Chg Pre-tax
Discretionary Distance YoY R&D YoY R&D YoY R&D YoY R&D income/
Year R&D (R&D) Q1 Q2 Q3 Q4 Revenue
2003 -0.010 0.146 -0.3% -3.3% 17.7% -31.7% 0.052
2004 0.102 0.187 17.7% 8.7% 25.7% -1.1% 0.009
2005 -0.019 0.092 8.2% 22.0% -17.9% 29.1% 0.000
2006 -0.010 0.294 18.4% 8.4% 24.5% 11.7% 0.022
2007 0.180 -0.634 16.3% 33.9% 52.8% 435.7% -0.133
WHARTON ONLINE
Company #3: Possible “Smoothing”
• NDOGTHA Inc.
Chg Pre-tax
Discretionary Distance YoY R&D YoY R&D YoY R&D YoY R&D income/
Year R&D (R&D) Q1 Q2 Q3 Q4 Revenue
2002 0.052 0.027 61.1% 68.8% 110.3% 30.5% -0.075
2003 0.009 -0.195 13.0% 14.7% -17.9% 25.5% -0.034
2004 0.030 0.367 31.1% 30.2% 14.8% 16.8% 0.021
2005 0.001 0.756 13.0% 2.0% 1.1% 11.1% 0.115
2006 0.082 0.935 40.3% 46.1% 58.4% 73.8% 0.058
WHARTON ONLINE
WHARTON ONLINE
