I keep my computers for a long time. So I don't want to do anything to shorten the CPU's life.
I would not worry too much about shortening the life of the CPU as long as you are not getting extreme with the voltage. For my last couple of systems I have been able to get "pretty good" overclocks by going with the philosophy of getting the best OC I can do while maintaining stability at stock voltage.
For example I've got one system that's a Core 2 Quad 6600 that I ran at 3.2 GHz when it was my main system (stock is 2.4 ghz). Others at the time were getting 3.6 by cranking up the voltage but I could get 3.2 at stock voltage. I bought that CPU in March of 2008 and it was my main system until May of 2011 (got a 2600K then). From May of 2011 until now, its been humming along 24x7 as a VMWare ESXi server, still at 3.2 Ghz (a 37.5% OC by the way). So that system is still going strong after 5 years of usage, with the last two being 24x7. I've had similar results with many other CPUs.
The point being that IMO, it seems like overclocking does not reduce CPU life in a practical way as long as the voltage is kept down. Sure, if I run that C2Q side by side with another one that is not overclocked and let them sit there running for years until they die, then the non-overclocked one, chances are would go longer. But other components like the motherboard are likely to fail a lot sooner. And if that socket 775 motherboard fails, it probably won't make sense to buy another one even if the CPU is still good. Thats why I say as a practical matter, overclocking, within reason, does not shorten the real world CPU life.