To ask the Secretary of State for Social Security if he will set out the percentage change in income during the period 1979 to 1987 both before and after housing costs of each percentile group in the population.
The figures are set out in the table. As explained to the hon. Member in my letter of 20 December 1990, these figures are not statistically valid; the differences between changes in consecutive percentiles are completely swamped by the large sampling errors involved. Both sampling errors and possible data problems at the extremes render the changes below the tenth percentile very uncertain, and particularly so below the fifth.Such statistics rest on a small number of cases, and are particularly vulnerable to distortion by household incomes which may be both exceptional and unrepresentative. The way in which the bottom percentile is distorted is clearly illustrated when comparing the position between the years 1983–85, which shows a 61 per cent. increase in real income for the bottom percentile.
Percentage change in real income by percentile of income distribution 1979–87 | ||
Percentile | Increase net of expenditure on housing Per cent. | Increase in income overall Per cent. |
1 | 1–22·6 | 1+6·7 |
2 | 1–16·2 | 1+6·8 |
3 | 1–7·3 | 1+10·4 |
4 | 1–1·8 | 1+9·9 |
5 | 1+0·1 | 1+8·5 |
6 | 1–-1·2 | 1+6·7 |
7 | 1–0·6 | 1+5·9 |
8 | 1–0·7 | 1+5·9 |
9 | 1–0·1 | 1+6·2 |
10 | 10·0 | 1+6·4 |
11 | +0·4 | +6·7 |
12 | +0·5 | +7·4 |
13 | +1·0 | +7·6 |
14 | +1·8 | + 7·3 |
15 | +2·4 | +6·8 |
16 | +2·6 | +7·5 |
17 | +3·0 | +7·9 |
18 | +3·6 | +7·8 |
19 | +3·2 | +7·8 |
20 | +3·0 | +8·0 |
21 | +2·8 | +7·9 |
22 | +3·1 | +8·2 |
23 | +3·0 | +8·6 |
24 | +2·5 | +8·6 |
25 | +2·8 | +8·6 |
26 | +3·4 | +8·4 |
27 | +3·6 | +9·0 |
28 | +4·1 | +8·8 |
29 | +4·7 | +9·4 |
30 | +4·9 | +9·9 |
31 | +5·3 | +10·1 |
32 | +5·6 | +10·3 |
33 | +5·9 | +10·6 |
34 | +7·2 | +10·6 |
35 | +7·9 | +10·6 |
36 | +8·1 | +10·5 |
37 | +8·8 | +11·0 |
38 | +8·9 | +11·3 |
39 | +9·7 | +11·4 |
40 | +9·8 | +12·0 |
41 | +10·9 | +12·1 |
42 | +10·7 | +12·1 |
43 | +11·1 | +12·4 |
44 | +11·3 | +12·8 |
45 | +11·9 | +13·1 |
46 | +11·8 | +13·5 |
47 | +12·0 | +14·0 |
48 | +12·9 | +14·1 |
49 | +13·6 | +14·3 |
50 | +13·7 | +14·3 |
51 | +14·0 | +14·4 |
52 | +14·8 | +15·1 |
53 | +15·0 | +15·2 |
54 | +15·1 | +15·6 |
55 | — | +160 |
56 | +15·6 | +16·2 |
57 | +15·4 | +16·2 |
58 | +15·7 | +16·4 |
59 | +16·1 | +15·9 |
60 | +16·3 | +16·0 |
61 | +16·5 | +15·9 |
62 | +16·7 | +15·8 |
63 | +16·7 | +15·9 |
64 | +16·9 | +16·4 |
65 | +17·5 | +16·7 |
66 | +17·8 | +17·3 |
67 | +18·2 | +17·7 |
68 | +18·5 | +18·0 |
Percentile
| Increase net of expenditure on housing Per cent.
| Increase in income overall Per cent.
|
69 | +18·9 | +18·0 |
70 | +19·7 | +18·3 |
71 | +20·0 | +18·7 |
72 | +20·4 | +18·9 |
73 | +20·7 | +18·8 |
74 | +21·2 | +19·2 |
75 | +21·4 | +19·3 |
76 | +21·5 | +20·5 |
77 | +22·1 | +20·9 |
78 | +22·8 | +21·6 |
79 | +23·6 | +22·0 |
80 | +24·4 | +22·3 |
81 | +24·8 | +22·9 |
82 | +25·2 | +23·3 |
83 | +24·7 | +23·5 |
84 | +24·5 | +24·2 |
85 | +25·0 | +23·7 |
86 | +27·0 | +25·7 |
87 | +27·9 | +26·5 |
88 | +28·3 | +27·2 |
89 | +29·3 | +29·0 |
90 | +29·5 | +29·5 |
91 | +30·3 | +31·0 |
92 | +32·2 | +32·2 |
93 | +34·6 | +33·1 |
94 | +39·1 | +35·4 |
95 | +40·3 | +38·8 |
96 | +42·5 | +40·6 |
97 | +49·0 | +42·3 |
98 | +55·2 | +50·8 |
99 | +72·4 | +61·5 |
1 Figures for the lowest decile, and particularly those for the lowest five percentiles, are subject to a very high degree of statistical uncertainty due to potential errors arising from the limited size of the sample and other data problems. |