Need clarification concerning bias parameter w0 in linear models for regression (Bishop textbook)

Can anyone explain what he means by "compensates"?

Thus the bias w0 compensates for the difference between the averages (over the training set) of the target values and the weighted sum of the averages of the basis function values.
-Bishop pg.142 (right after taking the log of the likelihood function and solving for w)

http://imgur.com/SVumkyl

What affect does the bias have on the model?

Thanks.