Regarding SSS declines:
According to the receipt method, SCP is down 17.5% from Q3 1997 (its opening quarter) to Q1 1998, or 9.2% per quarter. Let's assume, though, that it is decaying exponentially and not linearly to some value that will be reached at the end of 2 years, the time period which RAIN believes to be the end of SSS decline. A fit of these two data points to a three parameter exponential decay gives an intercept, or after tax profit, of 202,340 based on Q3 97 being 383,700 and Q1 98 being 316,900. This is the worst case scenario in my opinion as an exponential decay is unlikely.
Now, let's assume the decrease actually happens like this:
Q3 97 First open (383,700)
Q4 97 down 11% 341,493
Q1 98 down 7.3% (316,900)
Q2 98 down 5.8% 298,519
Q3 98 down 4.5% 285,086 SSS down 25%
Q4 98 down 3.4% 275,393 SSS down 19.6%
Q1 99 down 2.6% 268,233 SSS down 15%
Q2 99 down 1.4% 264,478 SSS down 11.5%
Q3 99 flat 264,478 SSS down 8.3%
Q4 99 flat 264,478 SSS down 4%
Q1 00 flat 264,478 SSS down 1.4%
Q2 00 flat 264,478 SSS flat
These are very crude approximations, but I believe they shed light on how misleading SSS figures really are!
Now, lets assume that we open 1 of each of this type of location every quarter and they all behave identically:
Quarter 1: 1 unit 383,700
Quarter 2: 2 units 725,193 (383,700+341,493)
Quarter 3: 3 units 1,042,093
Quarter 4: 4 units 1,340,612
Quarter 5: 5 units 1,625,698 SSS for 1 unit down 25%
Quarter 6: 6 units 1,901,091 SSS for 2 units down 23%
Quarter 7: 7 units 2,169,324 SSS for 3 units down 20.5%
etc.
So as you can see even though sales are actually leveling off, it takes an additional year for SSS to show this. Also, this model assumes all location open regularly. However, when a big "cluster" of locations opens in one quarter, it will depress the SSS significantly a year later.
So it seems to me that when we have so few locations open, SSS is not a meaningful statistic.
-Dylan |
