I have already explained this dynamic in a past article, but I feel it's something worth repeating as most investors would gain a lot from understanding it. When I wrote about Joy Global (JOY) being punished by coal (KOL) producer's travails, I mostly said that one man's capex is another man's revenues. However, there's an even deeper dynamic at work when we are talking about capacity sellers - that is, companies which sell the means for other companies to produce more.
Basically, capex is usually divided into investments that are needed to keep production running at its present levels, and more importantly, investments that are needed to expand capacity - to provide the ability to produce more. Now, the problem that happens here is that when an industry is working below its production capacity, there's little need to invest in additional capacity. So for those sellers whose main line of business is selling equipment that allows for expanded production, such as Joy Global, when this happens, revenues for a given segment can go from a high level down to close to zero in a very short while.
Let us say the coal sector needs to expand capacity in a given year by 50 million tons. Then in the next year, by 100 million tons, and in the third year there's overcapacity and no need to expanded output. This translates into equipment buys able to produce 50 million tons in the first year, 100 million in the second (a 100% growth rate, fantastic), and then suddenly in the third there's almost no need for new equipment except for wear and tear replacement, regular maintenance, etc.
This scenario makes for incredibly cyclical industries, one year being incredibly profitable and the next deep in the doldrums and losing money hand over fist. This is what can happen to Joy Global, and also in part to Caterpillar (CAT) or Terex (TEX). In the semiconductor business, investors are already used to this dynamic for the semiconductor equipment makers, so those companies get awarded low multiples at the top of their cycles. Also, semiconductor equipment gets obsolete fast, which helps with the cycles guaranteeing demand even if the semiconductor market sees a prolonged downturn. Not so with mining/construction equipment. This is thus a further negative for those expecting a quick turnaround for Joy Global.
As a seller of capacity, Joy Global is bound to see a much deeper and protracted correction than mostly anyone expects today. Even if coal recovers a good part of the demand it lost during 2012 due to dispatch switching - as I expect it to, the sector still mostly won't need expanded capacity as compared to 2011. This means orders from the coal sector are both bound to drop a lot and to stay low longer than expected (given coal's impending 2013 recovery).
In Joy Global's recent quarterly earnings, this fast-acting dynamic can already be seen in the plunging orders for mining equipment (-34% year-on-year). The cuts, however, are bound to be deeper.