Owning the chain from cotton through to a finished garment is one of those decisions where the right answer depends almost entirely on what is being made. For some products the investment repays itself within a few years. For others it never does. The discipline is to do the arithmetic honestly before committing, instead of assuming that owning more of the chain is always better than working with partners.
The arithmetic works when three things are true at once. The volume is steady enough that the owned stages run close to full, because an idle machine still costs money. The quality requirement is high enough that failures at a partner stage would cost more than the investment. And the pressure on lead times is real, because owning the chain shortens them in a way that customers will actually pay for.
It fails when volume is uneven, when the quality requirement is moderate, or when lead times are comfortable. Uneven volume leaves owned capacity standing idle with its costs unchanged. A moderate quality bar means partner failures are absorbable, which removes one of the main advantages. Comfortable lead times mean the saving cannot be sold. In any of those cases, working with partners is better on the numbers, and the numbers are right.
The middle path, which is owning the stages that decide quality and partnering for the rest, suits most businesses most of the time. Controlling dyeing and finishing while buying spinning and weaving captures much of the control without the cost of owning everything. It is harder to manage in the first year and usually cheaper across a decade.
There is a further point that rarely appears in the business case. Owning the chain concentrates risk in one place. A business with its own spinning, weaving and finishing is exposed to a single set of local conditions, from power supply to labour availability to weather. Partnering spreads that exposure, and the value of the spread only becomes obvious in the year something goes wrong.
Our own position is that of a buyer rather than an owner. We own no production, and some of the mills we buy from in India run integrated operations where their product and volumes justify it. What we bring is the ability to judge whether a mill's integration is genuinely serving the cloth or simply serving the mill's own utilisation, which are not the same thing and do not always point in the same direction.
If you are weighing this, start with the utilisation question and be pessimistic about it. Most business cases for integration assume demand that arrives smoothly, and most demand does not. A plan that works at seventy per cent of the forecast is a plan. A plan that only works at full forecast is a hope with machinery attached.
