This is the fundamental question that the experts have been asking themselves when it comes to interpreting the decline in birth rates. This question arises because the levels and trends in period fertility indicators, like the Total Fertility Rate (TFR), could be driven by either one of two mechanisms: changes in the number of children that women have, and changes in the timing of births. The TFR measures the average number of births a woman would have by the time she reaches the end of her reproductive years, if she experiences the age-specific fertility rates observed in a given period. This measure “transposes” a momentary experience across the lifetime of a cohort. The postponement of fertility introduces distortions into this transposition. The completed fertility rate (CFR), on the other hand, presents the actual average number of children that women of a real cohort do have by the end of their childbearing years. The changes in completed fertility are often referred to as “quantum” effects, while the changes in timing of childbearing are often called “tempo” effects. The bottom line is that because of the postponement of births, TFRs underestimate somewhat the completed fertility that will be reached by the cohorts currently in childbearing ages.
A brief examination of how this works in practice - based on the Swedish case - can be found here.
The classic exposition of the tempo and quantum issues as put forward by John Bongaarts and Griffith Feeney can be found here in presentation form, and here in paper form.