4 IPOs Debut, 3 Under Water
(06731/09607/03228/03757)

How the four debuts actually traded

Same listing day, four very different pictures. The table compares offer price, opening price and the 10am level. Per-lot profit and loss excludes commission, stamp duty and other trading costs.

Kinwong is the instructive one: it opened 7% below the offer price, then climbed from a HK$64.20 low to HK$73.30, a 4.9% gain. A weak open and a weak close are two different things, so judging the day from the first print is unreliable.

Does heavy oversubscription mean a sure first-day gain?

Subscription demand varied enormously across the four deals. Put the demand data next to the first-day outcomes and the “oversubscription equals win rate” idea falls apart:

From 207x down to 2.39x, first-day performance follows no order: the most oversubscribed rose, the least oversubscribed fell hardest, and the two in between split one up, one down. First-day pricing is set by several things at once — where the offer price sits relative to the A-share price, cornerstone investor participation, the international tranche, sector sentiment and how concentrated the free float is. Oversubscription captures only one of those (retail demand). Using it alone means guessing four variables from one number.

Cornerstones are another counter-example: Kinwong’s cornerstone investors took about 47.69% of the global offering and RoboTechnik’s about 35.2%, meaning a smaller free float that should, in theory, be easier to hold up. Both still came under pressure on day one. A high cornerstone ratio is not price support.

The expected value of a subscription: allotment rate x per-lot P&L

A subscription has two outcomes, allotted or not, so the number to look at is expected value: expected value = one-lot allotment rate x per-lot profit or loss. Using 10am prices:

Applying for one lot in each of the four, and assuming all four were allotted, ties up HK$35,704 in entry money and shows a paper loss of HK$1,694 at 10am prices, about -4.7%. Measured at the open the loss was HK$1,620. Neither figure includes commission, stamp duty or margin interest, so the real cost is slightly higher.

Red Avenue is the clean example: a 100% one-lot allotment rate means almost every applicant got a lot, and the stock fell 10.9%. Kinwong is the mirror image: only 3% of one-lot applicants were allotted, and even with a 4.9% gain the expected value works out to just +HK$10. Being allotted and being rewarded are different things — a high allotment rate usually signals that the market sees little upside left in the price.

Compare Transwarp (06727), which listed on 21 September: entry cost HK$4,949 per lot, a 10% one-lot allotment rate and a HK$62 loss per lot in the grey market, an expected value of about -HK$6 per application, close to break-even. Subscribing to IPOs is not a habit that always pays; it is a decision whose expected value can be negative.

Entry cost varies 10-fold: capital efficiency matters

The entry cost per lot differed hugely: Direct Drive Tech HK$2,181.78, Red Avenue HK$4,444.38, Kinwong HK$7,058.47 and RoboTechnik HK$22,019.85. One RoboTechnik lot ties up as much cash as ten Direct Drive Tech lots. The same “few hundred dollars” outcome uses very different amounts of capital:

• Direct Drive Tech: +HK$114 on HK$2,181.78 deployed, +5.2%.
• RoboTechnik: -HK$1,670 on HK$22,019.85 deployed, -7.6%.
• Red Avenue: -HK$480 on HK$4,444.38 deployed, -10.8%.

Expensive lots push many investors towards margin financing. That adds interest cost, and when the allotment rate is low, even a lucky allotment may not be large enough to cover the interest. Some channels also charge a subscription fee that is payable even when nothing is allotted, pushing expected value further down. So the question is not only “how much does a lot make” but “how much cash does it tie up, and how much of my available capital is that”.

A+H discounts of 30-40% are not simply upside

Three of the four are A+H listings (Red Avenue on the Shanghai exchange, 603650; Kinwong on Shanghai, 603228; RoboTechnik on Shenzhen’s ChiNext, 300757). Converting the A-share price into Hong Kong dollars at the day’s rate (HK$1.1696 per renminbi) and comparing with the H-share price:

A 30-40% H-share discount invites the thought that the Hong Kong line is cheap for the same business. In practice the discount is the norm for A+H companies: capital cannot move freely between the two markets, and the investor bases and valuation habits differ, so the two prices stay apart. Transwarp (06727), listed on 21 September, traded at a discount of roughly 63% to its A-share and still broke issue price on day one. A wide discount does not have to narrow, and H shares cannot be converted into A shares, so it is not an arbitrage.

The grey-market “fat finger”: grey-market prices are not real prices

Ahead of the listing, the grey market produced a scene worth remembering. Red Avenue was priced at HK$44.00, and shortly after the grey market opened on the Phillip platform it plunged to HK$0.985 — 97.76% below the offer price, a paper loss of HK$4,301.50 per 100-share lot. Phillip later confirmed that a very small volume did trade at that abnormal level.

The three grey-market venues eventually closed between HK$36.14 and HK$36.74, down 16.5% to 17.9%. Direct Drive Tech closed 3.4% to 4.4% higher in the grey market, Kinwong fell about 6.2% to 6.3%, and RoboTechnik closed at HK$428 on all three venues, down 1.83%.

Two practical lessons. First, grey-market liquidity is thin and prices can move far away from anything reasonable, so read volume alongside price. Second, investors who habitually park limit orders far below the market (“bottom-fishing” orders) should note that such prices can actually fill — anyone who bid HK$1 for Red Avenue transacted at a level they almost certainly did not intend.

Three questions before you subscribe

First: is the expected value positive or negative?Multiply the one-lot allotment rate by the per-lot profit or loss, then subtract fees and interest. If the answer is negative, the reason to subscribe is not “IPOs make money” but “I like this company after listing” — two completely different decisions.

Second: how much cash does one lot tie up? Work out the share of your available capital, whether margin is involved, and what the interest costs. A HK$20,000 lot and a HK$2,000 lot are risks of a different order.

Third: when do you plan to exit?If the plan is to sell on day one or in the grey market, accept that the price is largely random. If the plan is to hold, go back to fundamentals: revenue growth, gross margin, customer concentration and industry position — not the oversubscription multiple. FinKit’s IPO data pages carry the prospectus figures for these listings, so each item can be checked side by side.

FAQ

Q1: Is a heavily oversubscribed IPO a good one? No. Oversubscription measures demand, not pricing, float or fundamentals. In this batch the 207.56x deal rose and the 2.39x deal fell hardest, so there is no stable relationship.

Q2: Is a 100% one-lot allotment rate good or bad? It usually signals muted demand, meaning the price already reflects little upside. Red Avenue had a 100% one-lot allotment rate and fell the most on day one.

Q3: If the grey market falls, must the first day fall too? Not necessarily. Kinwong fell about 6% in the grey market and opened 7% below the offer price, then reversed to a 4.9% gain.

Q4: Are more IPOs coming? Yes. Camsense Technologies (06802) is expected to list on 30 September at a fixed offer price of HK$58.85, 100 shares per lot, an entry cost of about HK$5,944.35. It supplies spatial-perception components for robot vacuum cleaners. Running the same three questions beats following the crowd.

The bottom line

Four IPOs listed on one day and three sank, but their fundamentals, pricing and market reception all differed. The conclusion is not “IPOs no longer work” but “each one has to be worked out separately”. The three indicators most often quoted — oversubscription, grey-market price and the H-share discount — each produced a counter-example on the same day.

The practical approach is to reverse the order of the decision: work out expected value and capital tied up first, then look at fundamentals, and only then ask whether to subscribe. Do it the other way round and the subscription comes first, the reason afterwards — which leaves the outcome to luck.