Well, personally, I tried to escape by going and getting a Master's and a PhD. Unfortunately for me, I didn't bother to really LOOK at academia as a career path, so when I finished and started looking into it ... well, it's pretty crap, being an academic, no matter which country you're considering. Also, academia only considers the work you've done post-PhD as being of any worth, so I'd be competing with kids who'd gone straight through, rather than those who'd spent 20 years in industry first.
I've considered a wide variety of things, but always come back to this: if I keep on saving at my current rate, I can not have to make any profit whatsoever, whereas if I start out now with something (a bakery, say), then I'll need to keep on making a profit, so that I can retire some day. I've got another 10-15 years in software and then I'm done and can do whatever. Check out http://www.mrmoneymustache.com/2012/01/13/the-shockingly-sim... to figure out where you land on that scale.
That chart, I believe, does not take into account dividend reinvestment. I picked their worst datapoint on a twenty year horizon, '61-'81. I plugged January '61-January '81 into this calculator [1]. With CPI adjustment turned on, the annualized return with dividend reinvestment was 2.3%. That's the worst twenty year period in history.
Another difficulty with this analysis is that retirees do not withdraw lump sums twenty years after they retire. They withdraw smaller amounts each year. I have created a spreadsheet to simulate a retirement starting in 1961 [2]. You were probably feeling a bit nervous in '81, with your nest egg down to two thirds of its original size. But even in this terrible, terrible scenario, by following the 4% rule, you arrive at the same amount of real wealth you started with after thirty years. If you were lucky enough to live for forty years of retirement, you're two and a half times as wealthy as when you began.
Please let me know if you see any errors in my calculations. I've made notes on how I arrived at each number in the comments on the columns starting in year 1961.
The actual worst year for early retirement was '65, when a pure 4% rule portfolio would have failed after twenty five years. But, if you build some engineering tolerances into your spending plans, then even that was survivable [3]. Engineering tolerances in this case are meant to refer to leaving yourself room so that you can spend less if you must.
I've been following the Money Moustache type approach for a while now myself. I highly recommend it, but the closer I get to the goal the more I realise I need something else constructive to do if/when I jack it all in.
I've known a few people who've successfully retired using this approach. For me I don't see the point, who wants to spend their lives just saving every penny, they seem to get great satisfaction out of it though. Personally I've found a better way is to find work you enjoy and people will pay you for, find somewhere you like to live and enjoy the rest of your life without counting pennies on bars of soap.
I'm doing all of the above: living in a place so like, working at a job that is interesting and pays well. My new job has tripled my salary, and so rather than inflating my lifestyle, I've just been saving & investing the new money. I'm happy spending the same old amount based on my old salary, and just taking a couple more vacations a year.
I've considered a wide variety of things, but always come back to this: if I keep on saving at my current rate, I can not have to make any profit whatsoever, whereas if I start out now with something (a bakery, say), then I'll need to keep on making a profit, so that I can retire some day. I've got another 10-15 years in software and then I'm done and can do whatever. Check out http://www.mrmoneymustache.com/2012/01/13/the-shockingly-sim... to figure out where you land on that scale.