Cyclical PhD production

I was wondering why recently nobody graduated with a PhD from our lab. My first (and wrong) conclusion was, that our lab is not as productive anymore as it used to be. When I looked at the data carefully I discovered an interesting phenomenon: The PhD production rate in our lab follows a 5-6 year cycle. And lo and behold: The cycle is about to peak again in 2013/2014, which is exactly what I expect to happen as many of our PhD students are about to finnish this or next year. I wonder whether this is specific for Kari Alitalo's lab or whether this a general phenomenon. In any case I am looking forward to all those parties ("karonkka") that lie ahead.
The figure above shows the number of PhDs received any given year that were supervised by Kari Alitalo averaged over a window of one or three years. Shared supervision was counted as 0.5.