A seven-month stall at Hyundai Mobis shows what Korea's union law does to carve-out timetables

What a bidder actually underwrites here is the calendar. Seven months of stall on the Mobis lamp contract, counted back from the 30 August account ^[3][5], puts the signing block at roughly late January ^[18], and if you charge the unreceived proceeds a 10 per cent annual cost of capital, the wait has cost about 5.4 per cent of their present value ^[20] (my rate, not anyone's disclosure), which is a deliberately unimpressive number: small enough to bridge with price, awkward enough to argue over for another seven months. Nor is it one file. At Kyungdong City Gas, a customer service centre union ran a relay hunger strike in July against the prime contractor's planned sale of a subsidiary stake, on the grounds that it could lead to outsourcing ^[4].

Since the revised act took effect, unions have carried matters once handled as management prerogatives into bargaining and industrial action ^[1], and the delay is manufactured by where the boundary sits rather than by any explicit veto: each dispute turns on an assessment of how substantially and concretely a management decision touches working conditions ^[16], so the answer arrives after announcement, from a tribunal or an inspector. An employment covenant does not close that gap on its own, as the Mobis file demonstrates.

The counter-thesis lives in the two decisions already handed down, and I think it is at least as strong as mine. In the Welliv dispute the National Labor Relations Commission recognised Hanwha Ocean as an employer for matters such as industrial safety and working conditions, but declined to make prime-contractor-level performance bonuses a mandatory bargaining subject ^[7][8]; at the National Tax Service, where outsourced call centre workers asked for direct hiring and a reworked wage structure ^[9], the employer status the labour ministry recognised covered the work environment and protections for emotionally demanding roles, after which the demands widened to direct hiring anyway ^[10]. Both stopped at conditions and left pay and employment status outside ^[19]. Read that way, the law lengthens deals instead of stopping them, and length is something a discount rate handles.

There are three ways this resolves, and they differ by who does the resolving. The guidance publishes, both sides work from a template, and the Mobis contract lands near its agreed terms ^[14], which would say the ambiguity rather than the statute was binding. Or a tribunal reads substantial effect widely, prior consent to asset sales becomes bargainable in fact, and the line Park Ji-soon of Korea University Law School describes, the exclusive authority of the board and the shareholders' meeting ^[13], starts to move. Or the politics settles it, President Lee Jae-myung having already told officials that the scope of industrial action looked too broad and asked for clear standards ^[17], with the caveat that an instruction is not law either.

My view, held loosely: Korean carve-outs trade a few points wider on timing rather than at a structural discount, because only the delaying kind of objection has been adjudicated so far ^[19]. What would prove it wrong is a labour-heavy Korean unit repriced or withdrawn outright over union objection. Meanwhile the bill sits with the seller, which has spent the better part of a year running a division it has agreed to hand over, proceeds neither banked nor spent ^[5].