On cycles  
  
A cycle means a repetitive pattern with a reasonably recognizable   
periodicity, and should not be confused with a random walk.  
  
I had cause (long ago) to look at data for C19 Britain. My memory is that   
the early C19 did not show recognizable cycles - variations, of course,   
good and bad times, financial crises, etc., but no recognizable   
periodicity. Writers of the time did not write about cycles not because   
they failed to notice them, but because there weren't any to recognize.  
  
By contrast, the second half of the C19, for Britain (but not necessarily   
other places) had very visible 7-10 year cycles, which really jump out at   
you from all sorts of data series, and must have been obvious at the time.   
Hence, late C19 economists in Britain (which dominated the subject at the   
time) treated trade cycles as an important topic.  
  
For the C20 (at least after 1914) cycles are much less clear. The first   
half was dominated by 2 wars and the great depression of the 1930s. The   
second half had a quite different pattern, with shorter and more variable   
cycles (if they were properly called cycles at all), plus one-off events   
like the oil crisis. Government policy almost certainly played a much more   
important role, so what cycles there have been in the later C20 are   
probably different animals from C19 cycles. Without late C19 Britain, we   
might perhaps not have a recognized topic of trade cycles.  
  
Actually, the clear cycles of late C19 Britain are less straightforward   
than they look, since home and foreign investment followed out-of-step   
20(ish) year cycles, with successive cycle peaks dominated by home and by   
foreign investment.  
  
Tony Brewer