Research
Return to News Page
OGI Test Methodology
The model was designed to be conservative. High hurdles were
set where applicable in order to increase the probability of finding
companies with substantial organic growth. The Organic Growth
Index model consists of seven different steps or tests. The first
three tests are designed to illuminate the fastest-growing companies.
The last four tests are designed to illuminate companies with
the most organic growth among the pool.
Test 1: Economic Value Added/Capital
We started with the EVA Dimensions Performance Rank 1,000 EVA/MVA
model as a market-accepted definition of economic value creation.
For each period, we took the top 1,000 EVA companies for the base
year. For accounting reasons, we eliminated real estate investment
trusts, insurance companies, and financial companies. We then
computed EVA/capital invested (equity + debt) for each company
for each year, ranked them, and then averaged the ranks across
the applicable time period. We chose the top 300 companies as
"winners."
Tests 2 and 3: High Growth Companies
How many of these 300 companies were high-growth companies as
compared to their industry competition? Was growth better evidenced
by increasing revenues or by increasing cash flow from operations
(CFFO)? We decided to use both tests and weight them equally.
Test 2-Sales CAGR. For each of the 300 companies in each test
period, we computed their sales compounded annual growth rate
(CAGR) and compared it to a six-digit industry average of sales
CAGR using over 6,000 different companies. We eliminated outliers
(i.e., any company that was 3 or more standard deviations from
the industry average). We then computed a Z statistic for each
company to normalize our results across industries.
Test 3-CFFO growth. For each of the 300 companies in each test
period we computed the annual change in cash flow from operations.
We then divided the annual cash-flow change by the book value
of assets in the initial year in order to have a ratio of the
magnitude of the change. The results for each company were averaged
across years and compared to the six-digit Global Industry Classification
Standard average change in cash flow from operations. We again
used Z statistics to normalize results across industries.
For test 2 and 3 we used data from S&P's Compustat database.
To get the end result we weighted the Z results of tests 2 and
3 equally and averaged them. If a company had a positive net Z
statistic, it passed these two tests.
Test 4: Modified S&P Core Earnings Test
S&P defines core earnings as "income associated with
a company's ongoing operation." In our model, its test was
modified by dividing a company's average S&P core earnings
for the applicable test period by its average reported net income
to create a percentage.
If the percentage was greater than or equal to 90%, a company
passed the test. The underlying assumptions of this test were
that core earnings were a proxy for organic growth and that the
higher the percentage of core earnings as compared to reported
net income the higher the likelihood a company was growing organically
and not by creating one-time nonrecurring earnings.
Test 5: income Manipulation
A common revenue manipulation is to accelerate income recognition
either through changing income-recognition policies, extending
more liberal credit terms, or channel stuffing. One commonly used
method of illuminating this possibility is to compare the growth
rate of receivables to the growth rate of sales. If receivables
growth outpaces sales growth, something unusual may be going on.
We looked at each company's year-by-year change in receivables
and sales and averaged their changes for the applicable time period.
If the average annual growth rate of receivables grew more than
10% of the annual average growth rate in sales, the company flunked
this test.
In looking at the results for test 5 we noticed an anomaly.
Some companies flunked this test even if they had a de minimis
amount of accounts receivable. This unintended consequence was
dealt with by adding a de minimis exception: companies with overall
accounts receivable less than 5% of sales automatically passed
this test.
Test 6: Merrill Lynch's Cash Realization Test
One of Merrill Lynch's four analysis screens in its 2002 Quality
of Earnings Report looked at the ratio of a company's CFFO to
its net income. For each company entering test 6 for each year
of the applicable test period we computed this ratio and then
averaged the ratios. If the average was equal to or greater than
90%, a company passed the test. For this test and test 4 we created
special-decision rules to apply if a number in any year was a
negative number.
Test 7: Mergers and Acquisitions Test
Historically, academics define organic growth as non-acquisitive
growth. Again, our purpose was to illuminate high organic growers,
not serial acquirers of revenue. This was a difficult test to
construct because of unreliable data on the amount of income acquired
with a deal. Because of this, we used deal values from the Securities
Data Corporation's database and assumed that deal values were
a proxy of income acquired, assuming accretive deals. For each
test period, we added the sum of deal values for each company
as the numerator and divided that sum by the company's increase
in market cap over the same time period (defining market cap as
equity plus debt). If the resulting ratio was greater than or
equal to 35%, the company failed.
|
|
|
| |
|
|
|
Copyright © 2008
