Anyone who would dare buy Hewlett-Packard’s shares must take four points as given. One: HP is badly managed. Sure, Meg Whitman got the top job after the acquisition of Autonomy. But would it be prudent to expect astute decision-making from HP’s leaders, given recent events? Two: four of HP’s five operating divisions (personal computers, printers, services, enterprise hardware) are facing secular declines in revenues, margin compression, or both. Again, the historical record and the performance of HP’s competitors make it foolhardy to bet any other way. Three: HP’s growing software business, is a bit, ah, hard to predict. After the Autonomy debacle, does anyone want to argue this? Four: HP’s revenue and cash flow are most likely in permanent decline. This last point follows from the first three.

In the fiscal year just ended, HP generated a shade under $7bn in free cash flow. Assume that declines at a steady 10 per cent a year into perpetuity. The present value of those future flows, at a middling discount rate, is about $20 per HP share. Back out the company’s net debt, and a figure very close to the company’s $12 share price results. If the decline is slower than that, the shares are cheap, in theory. At a 5 per cent rate of decline, on this wildly simple model, the stock is worth $21.

The problem is that real world declines, especially in technology, are never steady. The ups and downs that characterise struggling businesses scare investors beyond the point of thinking about their share of future cash flows. The possibility of a near-term death spiral is always priced in. This is why so few declining businesses have been successfully “run for cash” – even when, in theory, that is the best option. In short, buying HP shares is for gamblers only. Investors should stay clear.