Do you have evidence that this is the case? If not it seems like a very big assumption.I usually presume ages on the censuses are rounded up a year for the most part.
In my experience the ages of young children, servants, and very old people are more likely than most to be overstated, but there are plenty of people whose age is understated, so if you presume their age has been rounded up you'll be at least 2 years out.
