Comparisons - "current versus reformed" & various cases
Reasoning for the presentation style used here
Comparisons based on the number of qualifying children
Comparisons based on housing costs
Marginal increases in liability as net income increases
Marginal increases in total deductions as gross income increases
A related topic - NRP incomes and the new formula
Home & weblog
Blog archive & site history
Site map & search

Reasoning for the presentation style used here


The aim in this section of the web site is to be able to make generally-applicable statements and draw generally-applicable conclusions about the reformed system in order to judge what the impact of the reforms is likely to be.

The aim is not to provide a prediction for any particular person about how s/he will be affected by the reformed system. This site has an assessment calculator (in the form of an Excel spreadsheet) which gives a readout of both current and reformed liabilities, and that is much more useful for people wanting to check their own circumstances. Other assessment calculators are available too, and links to them are available here too.

Presentation options

All comparisons here are of the form: "for people in a given set of circumstances, these are the liabilities for different incomes". These are presented as graphs. 3 forms were considered for use throughout this site. In each case on this page, only the presentation format changes - the circumstances are the same, they are just different ways of displaying identical data.

1 - Absolute liability versus net income

Circumstances: 1 young qualifying child, housing costs of £50 per week, no other complications.

The vertical axis is the liability, the horizontal axis is the net income.

This is useful for an individual: "how much will I pay per week?" But assessment calculators are much better tools for providing that information, and enable a number of "what ifs" to be examined.

Obviously the amount of money that someone thinks significant depends on net income. £5 means far more to some earning £100 than someone earning £1000. So the £40 (about) difference shown over much of the graph between current & reformed schemes means more to a person earning £300 than someone earning £700. (Indeed, research shows that £5 means somewhat more to someone earning £100 than £50 means to someone earning £1000).

This format underestimates the importance of the differences at low incomes and overestimates the importance at high incomes. This is not the chosen format.

2 - Remainder of net income after payment versus net income

Circumstances: 1 young qualifying child, housing costs of £50 per week, no other complications.

The vertical axis is the amount of net income left after paying the liability.

The amount paid is obviously of interest, both to the PWC and to people wondering "do children really cost that much?", see the previous graph. But what matters most to the NRP is how much s/he has to spend on everything else afterwards. If you pay out lots but still have lots left, you are OK. If you only pay out only a little yet have nothing left, because you didn't much in the first place, you starve.

So this version shows how much the NRP has left of the net income after paying child support. This sounds fine in principle, but it is non-intuitive, and it suffers from the same problems as format 1 above by making an amount appear equally significant at all net incomes. This is not the chosen format.

3 - Percentage of net income paid versus net income

Circumstances: 1 young qualifying child, housing costs of £50 per week, no other complications.

This format shows the percentage of net income paid in the vertical axis.

For an individual is this is actually less user-friendly that the other formats. It doesn't even talk money! (Although a simple sum will yield the same answer as either of the other 2 formats.

But for analysis & explanation purposes this shows important things at a glance. This is useful for policy-making purposes. Just see what leaps out immediately:

  • differences at low incomes are put into better perspective - the £40 (about) difference at £300 is seen to be more important than the same difference at £700
  • and see that steep rise in the current formula between £150 and £250, where a modest rise in income triggers a massive relative rise in liability (and the reverse, of course!)

Although this can exaggerate the fine detail at low incomes (for example, that peak at £50 is exaggerated by only doing the sums at £50 intervals, and is about £5 in each case), this appears to be the best compromise. This is the chosen format.

Disclaimer about net income

Unfortunately, "net income" means different things under the current and reformed schemes. (Under the current scheme, only half of the pension contributions are subtracted from gross income, while under the reformed scheme all of the pension contributions are subtracted).

Therefore, some NRPs will have a lower net income under the reformed scheme than the current scheme. The difference can at most be a few percent, because of the rules for pension contributions, but it is still a problem that I am unable to correct for. What it means is that if the NRPs' net incomes are calculated according to the current scheme rules, sometimes they will pay slightly less under the reformed scheme than these graphs indicate.

Example 1. Consider an NRP with 1 qualifying child and a net income of £250 under the current rules. (This person is near the peak on these graphs - about 28%). Under the reformed rules, that person will, as shown, pay 15% - but if the NRP pays into a pension, it will be 15% of a slightly smaller amount. (This would normally, but not always, be equivalent to paying a smaller percentage of the same amount. For 1 qualifying child for a person paying the maximum allowed into a pension fund this could be equivalent to paying slightly less than 14%, instead of the standard 15%, and for 3 qualifying children paying slightly more than 23% instead of the standard 25%).

Examples 2. Consider an NRP also with 1 qualifying child but with a net income of £200 under the current rules. (That person is seen to be paying about about 16% under the current scheme). At £200, that person would at first sight also pay 15% under the reformed scheme. But in fact, s/he will be earning less than £200 under the new rules, and hence pay less than 15%, so will pay a lower percentage of a lower income under the reformed scheme. The difference for 1 qualifying child for a person paying the maximum allowed into a pension fund could be that instead of paying 15% of £200 (£30), s/he pays about 14% of about £185 (£26). For 3 qualifying children it could mean the difference between paying £50 and £43 in child support.

These differences are quite small at the resolution of these graphs, and don't matter much for the sort of broad comparisons being made here. I suspect that few NRPs pay the maximum pension contributions anyway. (The comparisons here anyway don't take into account inflation and wage increases before the reformed scheme is implemented). The calculator available at this site takes the pension rules into account by making the pension contribution explicit, and it is a much better way for individuals to compare their liabilities under the different rules - it gives a direct readout of the amounts for the two schemes.

What may be more important is that the change in rules will encourage NRPs to pay more into pension funds! Is this a deliberate bit of social engineering? If the government has made the decision "let's encourage NRPs to pay more into pension funds, then set the percentages so that they deliver the intended amounts of child support if the NRP does pay into a pension fund" - then I approve!

Doing the calculations

The assessments are produced using the assessment spreadsheet available at this site, and transcribed as series-data into another spreadsheet where the graphs are drawn using standard Excel facilities. (They were then pasted into Paint Shop Pro version 5 to make them suitable for current purposes).

Anyone is welcome to produce more comparisons & analysis and submit the results to this site as a guest article.

"First draw the curve, then plot the data."

Page last updated: 4 July, 2004 © Copyright Barry Pearson 2003