Four megacaps added 300 S&P 500 points in Q3 while everything else gave back 150

Citadel Securities' 300-point figure, as quoted in HORAN Capital Advisors' letter, and the letter's own sector figures are not measured the same way, so the 200% multiple should be treated as an order of magnitude ^[1]. The four stocks' 300 points against the roughly 150 lost by the rest of the index nets to about 150 points of index gain ^[2][13]. If that is the quarter's 2.30% total return ^[5], 300 points would be worth about 4.6 percentage points ^[19], and the letter credits only 2.72 to the entire technology sector ^[5]. One count is in index points and the other in total-return percentage points. On the sector count alone, technology's 2.72 points are 118% of the 2.30% return ^[14].

The case against alarm is in the same letter. HORAN cites final second-quarter earnings growth of 53.7% year on year from LSEG I/B/E/S and expects 30.6% for the third quarter ^[9][10], and says the underlying economy "held up well" ^[11]. The top-ten weight ^[3] pays off only if those ten companies produce that growth. The earnings figures are not broken out by company.

The four keep earning and the weights are vindicated, which the earnings growth above is consistent with. Rotation resumes, as it did while the equal-weighted index led until September ^[6], and the gap closes without the index falling. Or the four stall while the rest stays weak, and the cushion goes: HORAN wrote that the large mega cap technology stocks "have masked the weakness in the market" ^[18].

In our view the concentration is real and the risk is conditional. The loss arrives in a quarter when the four fall and the rest of the market, already weak underneath an index less than two points from its high ^[8], does not offset them. That holders are unaware of these weights is our assumption; the source measures the weights themselves ^[12]. We would be wrong if the equal-weighted index beats the cap-weighted one for a full quarter while the four contribute less than the rest of the index, because that would show the breadth is real and the Q3 tally was a one-quarter artifact.