CSIQ Shorts Retreat but the Stock Keeps Falling

Canadian Solar is caught in an unusual bind: short sellers are actually pulling back, yet the stock is down more than 21% on the week, a gap that points to something more structural than a bear raid.

The short-interest picture has shifted quickly. Positions fell roughly 13% in a single session on June 9, bringing SI down to 21.8% of the free float — still elevated in absolute terms, but the direction of travel has clearly reversed from the late-May peak above 26%. Borrow remains cheap at 1.52%, and availability has loosened to around 59%, a meaningful improvement from mid-May when it was below 35%. That combination — falling short interest, easier borrow, and a stock nonetheless shedding a fifth of its value in a week — suggests the selling is coming from long holders exiting, not from fresh bearish conviction in the lending market. The ORTEX short score eased slightly to 73.1 from 75.2 a week ago, consistent with a modest reduction in bearish positioning rather than any broader short squeeze.

The sell-side had already been ratcheting expectations lower before this week's slide. Multiple brokers cut targets sharply in late March — Roth Capital slashed its target from $30 to $15, Oppenheimer halved its view from $38 to $19 — and Freedom Broker downgraded to Hold in May with a $16 target. That $16 level is now essentially where the stock trades. Wells Fargo trimmed its Equal-Weight target from $23 to $17 in April. The direction of travel across the Street has been uniformly downward, with most firms parking ratings at neutral or equivalent rather than abandoning coverage outright. The mean analyst target is $17.19, only marginally above the current $16.12 print — a historically narrow implied return that reflects how far expectations have already been cut. Factor scores add nuance: earnings momentum ranks in the 98th percentile on a 90-day basis, meaning estimate revisions have actually been running positive even as the stock has collapsed, but the short score rank (2nd percentile) and days-to-cover rank (4th percentile) signal that the market's bearish framing hasn't fully unwound.

Options traders aren't particularly alarmed. The put/call ratio is running at 0.27, slightly below its 20-day average and well off its 52-week high of 0.69, suggesting there is no rush to buy downside protection at current levels. That reads as either complacency or a market that has already priced the bad news — the stock has fallen 20% in a month, so the options crowd may simply see less marginal reason to hedge.

The earnings calendar sets the next hard reference point. The most recent print on May 14 produced a 10.8% one-day drop and a nearly 10% five-day loss — a pattern that illustrates how badly the market has been punishing any disappointment. The next release is expected August 14. The stock's price-to-book multiple has compressed to 0.32, and the EV/EBITDA reads 9.9 — multiples that look undemanding if the company can stabilise margins, but which offer little protection if the solar pricing environment or tariff backdrop deteriorates further. Peer pressure is also acute: FSLR fell 15.7% on the week and JKS dropped 19.6%, confirming that the selling is sector-wide rather than idiosyncratic to Canadian Solar, though CSIQ's decline still leads the group.

The August earnings print now becomes the focal point — specifically whether the company can demonstrate that the estimate-revision momentum captured in the factor scores translates into actual reported results rather than another negative surprise.