This latter result does not coincide with Thoresen and Alstadsæter’s (2008), referred to
earlier, who found that business owners benefitted from organizational moves. Therefore, in order to
assess the effect of employing income measures for longer time periods, which would align the present
approach with that followed by Thoresen and Alstadsæter (2008), incomes are aggregated over five-
year periods (letting occupation of the last year of each period define occupational categories).10 As
shown in Table 8, organizational shifts now go together with upward movement in income rankings
for five-year income measures. There may be various explanations for the dependency on the
timeframe. The five-year specification, compared to the calendar year approach, captures lagged
effects of organizational shifts. “Smoothing” of incomes over time may also affect business owners
and wage earners differently, given the wider income variation among the former group (as seen in
Table 3). Five-year aggregations may therefore moderate the effect of one or two less profitable years.
We return to this periodization issue in the next section.
Table 8. Shifts from wage earner to owner of small business and shift from self-employment
and closely held firm to widely held firm. Percentage of “shifters” for combinations of
period t-1 and period t quintiles. Transitions between five-year periods, 1993-1998
and 1998-2003.
|
From wage earner to owner of small business |
From self-employment and closely held firm to | ||||||||||||
|
(29,468 obs) | |||||||||||||
|
From |
To quintile |
From |
To quintile | ||||||||||
|
quintile |
1 |
2 |
3 |
4 |
5 |
Total |
quintile |
1 |
2 |
3 |
4 |
5 |
Total |
|
1 |
13.9 |
3.4 |
1.5 |
0.7 |
0.6 |
20.1 |
1 |
6.6 |
2.8 |
1.9 |
1.2 |
1.8 |
14.3 |
|
2 |
5.2 |
5.9 |
3.7 |
1.5 |
0.6 |
16.9 |
2 |
1.5 |
3.1 |
2.3 |
1.9 |
1.8 |
10.5 |
|
3 |
2.4 |
4.0 |
5.9 |
4.5 |
1.3 |
18.1 |
3 |
0.6 |
1.4 |
3.1 |
3.1 |
2.7 |
11.0 |
|
4 |
1.1 |
2.0 |
4.2 |
8.5 |
4.3 |
20.2 |
4 |
0.4 |
1.0 |
2.4 |
5.7 |
6.3 |
15.7 |
|
5 |
0.3 |
0.6 |
1.0 |
4.0 |
18.9 |
24.8 |
5 |
0.3 |
0.6 |
1.3 |
5.5 |
40.7 |
48.4 |
|
Total |
22.8 |
15.9 |
16.3 |
19.2 |
25.7 |
100.0 |
Total |
9.3 |
9.0 |
11.0 |
17.4 |
53.3 |
100.0 |
4. Further description of the correlation between business
ownership and income movements
4.1 Specification of a transition equation
The relationship between business ownership and position in the income hierarchy is further explored
by a multivariate approach. The main advantage of a multivariate approach is that other observed
10 The empirical approach in Thoresen and Alstadsæter (2008) is also based on aggregated incomes, but not the same
aggregation as used here.
18