The statistical file associated with this article can be found in this reference.
The 2006 Census showed a break in the upward trend in the number of persons who gave their main occupation as professional musicians. The initial release, in May 2007, was a special compilation of cultural and leisure occupations. The number of other arts professionals including persons whose main job in the week before the Census were as actors, dancers, authors, visual artists and craftworkers also declined between 2001 and 2006.
Discussions with the National Centre for Culture and Recreation Statistics of the Australian Bureau of Statistics indicate that there has been no significant change in the way the Census is conducted. The Centre in December 2007 provided us with comparative statistics for 1996, 2001 and 2006 which largely confirm the trends shown in the May 2007 publication. (There are some changes because the latter was based on a revised occupational classification, ANZSCO, whereas the comparative table of December uses the previous classification, ASCO).
It is understood that the ABS intends to publish details of the culture-leisure occupational classification in February 2008, which we hope will enable a full analysis of the long-term picture which is the subject of this paper. Meanwhile we refer to the analysis of the key data from the 2006 Census which follows this paper.
Until additional data are published from the 2006 Census, as indicated in the previous section, this paper analyses trends up to and including 2001.
The occupational classification in Australia’s five-yearly Census of Population and Housing includes nine different categories of professional persons engaged in artistic and other cultural pursuits. All but one (photographers) have sub-categories, including musicians and related professions with separate statistics for music directors, singers, instrumental musicians, and composers. This article includes statistical comparison of the nine groups of arts professionals.
The Census records the main occupation during the week preceding the Census. Musicians are often employed in a variety of areas such as music teaching or administration, or their main paid job is outside the music sector. These persons are not counted as professional musicians in the Census.
Despite these limitations in coverage, the Census provides valuable data on professional musicians. Some may alternate between main occupations during the year, mainly teaching one week and mainly performing during another week, so the true number of full-time music professionals may be slightly underestimated. The large number of popular and other musicians performing gigs on a part-time basis are not included in the Census. These issues are discussed at length elsewhere in the knowledge base.
The story that follows refers to the tables in the attached statistical file. The story itself is illustrated by graphs rather than tables (click thumbnails to see in larger size, or inspect the better-produced originals in the statistical file).
But see the note on the 2006 Census, which has interrupted the long-term trend. As indicated in the introductory section, we are awaiting the detailed new statistics before integrating the new results into the story.
Table 1 shows the longest available time series for number of musicians counted in the Census (1961-2001). The footnotes to the table show this was made possible through research for three Australia Council reports, published between 1987 and 2000. This helped sort out discontinuities due to classification changes, except that musicians and related workers included private music teachers up to 1981, and reliable estimation of who was in which group could not be carried out except for 1981.
Chart 1 shows total numbers from 1961 to 1981 and the two groups separately from 1981 to 2001. Over the total forty years, the number of musicians and private music teachers increased almost fourfold from less than 3,700 to 17,400 (+375%). Over the twenty years from 1981 to 2001, the number of musicians and related workers increased by 88% from 4,800 to just over 9,000. This was less than the growth rate for private music teachers, from 3,725 to 8,441 (+116%). Average compound annual growth over the twenty years to 2001 was 3.0% for musicians and 4.3% for teachers; 3.6% for the combined group. Over the full forty-year period, the annual growth rate was 3.9%.
While there has been some slowdown since 1981, the rate of growth in the number of musicians significantly exceeded that of the total employed work-force, which grew by an average of about 1.5% (Table 2). However, the growth rate in the number of musicians has been consistently below that of the total number of arts professionals (Chart 2).
The data in this part are from Table 3 in the statistical file reference. The musicians and related professions group (just over 9,000 persons in 2001) was dominated by instrumental musicians (6,625 or 73.5%), followed by singers (15%). The smaller groups were music directors (4.4%), composers (3.7%), and those who couldn’t specify their occupation within the musicians groups (3.4%).
Most musicians are male according to the 2001 Census (71%). The proportion was highest among composers (83%) followed by instrumental musicians (77%) and music directors (59%). Only among singers was there a minority of males (47%). The male ratio in the total employed workforce in Australia in 2001 was 55% according to the main Census tabulations in Working Population Profile.
One Census (1991) distinguished between concert or opera singers and popular singers. The latter represented 74% of all singers counted in that Census. One may guess that a similar ratio existed among instrumental musicians, and that ‘popular’ musicians if counted separately would still outnumber ‘classical’ musicians by a large ratio.
The oldest persons among musicians and related workers were music directors and composers. The former had a median age of 40.9 years and the latter 40.3 years. Persons aged under 40 comprised 47% of music directors and 43% of composers, compared with 59% of all musicians and related workers. The median age for both instrumental musicians and singers was about 36 years, and 62% of singers and 60% of instrumentalists were aged under 40. The proportion of the total Australian workforce under 40 was 53%.
Only 60 musicians in the 2001 Census described themselves as Indigenous (0.7%). Though up from 44 persons in 1996, the proportion remained much below the participation rate for Indigenous people in the total employed workforce (1.2%).
Another 15.1% were born in other English-speaking countries and 8.7% in other countries, leaving non-Indigenous people born in Australia with a massive 75.5% (above the 73% found for the total employed workforce).
In summary, compared with Australia’s total employed workforce in the 2001 Census:
Table 4 compares gross weekly incomes for four categories of musicians, and those whose music occupation was unspecified or insufficiently defined. In 2001, the median gross weekly income for the whole group was $444, equivalent to an annual income of about $23,100 at the then prevailing prices (assuming the week preceding the Census was average for the year). The highest median of $777 was for music directors ($40,400 on an annual basis), followed by composers ($31,100), singers ($22,800) and instrumental musicians ($22,400).
The income distributions reflected this pattern with 37% of music directors earning $1,000 or more per week, and 25% of composers, but only 9% of singers and instrumentalists. At the lower extreme, 11-12% of music directors and composers earned less than $200 per week compared with 18-19% of singers and instrumental musicians.
Professional people whose main job was music were the second-lowest paid among the nine specified arts professions (with a median of $444 per week or $23,100 on an annual basis, as shown in the first paragraph of this section). Only visual arts and crafts professionals were paid less, with a median of $352 ($18,300). The highest-paid arts professionals were journalists and related workers (a median annual income of $44,800) followed by film, TV, radio and stage directors ($44,000), designers and illustrators ($34,600), authors and related professions ($32,700), media presenters ($29,800), photographers ($28,500), and actors and dancers ($24,750).
The median annual income for all artistic professions was $32,350, 40% above the median for musicians and related workers and above the median of all the subgroups of that profession except music directors. The median for all cultural workers, as defined by the Australian Bureau of Statistics and further discussed in the last section of this article, was almost as high ($31,600), and the median for the total labour force not far below at about $31,050.
The income distributions enable us to estimate the total income of arts professionals in “2001”, based on the typical week in August covered by the Census (Table 5). The first step is to estimate average incomes, which typically exceed median incomes by 20-25% as shown by the average/median ratio in Table 5. This is due to a long ‘tail’ into the high end of the income distribution.
The ranking of average incomes among the four categories of musicians and related workers, and among the nine categories of arts professions, remains the same as for median incomes. However, there are some significant differences in the ratio of average to median income. A relatively low average/median ratio signifies a relatively equitable income distribution. The ratio for all arts professionals was 1.22, that is, the average income exceeded the median by 22%. It was 1.20 for all cultural occupations (the concept further discussed in the last section of this article), and 1.23 for the total labour force.
Basically, income distributions are more equitable (relatively low average/median ratios) when the occupational composition of the group in question is reasonably homogenous. There is an opposite force, however, when many highly-paid ‘star’ performers emerge in a particular occupational group.
The relatively lowly paid private music teachers had an average ratio of 1.23, indicating inequalities about equal to the total labour force though at a much lower level: the median and average for private music teachers were both 42% below the median and average for the total labour force.
The total group of musicians and related workers showed relatively high inequalities, with an overall average/median ratio of 1.29, but music directors’ incomes were relatively more equitable (1.21) as well as being the highest among musicians with an average income of almost $48,800.
Composers’ incomes were the least equitable in the musicians’ group suggesting there are some highly paid individuals pushing the average up to 30% above the median income. Instrumental musicians’ incomes appeared to be less equitable (the average/median ratio was 1.28) than singers’ incomes (1.25).
Among the different categories of arts professionals, the least equitable income distributions were for media presenters (1.31), musicians (1.29), visual artists (1.28), and actors and dancers (1.27). All these groups have well-known top performers who presumably command relatively high incomes. Authors with an average/median ratio of 1.22 don’t seem to have nearly as long a high-income tail on the income distribution. Their ratio is closer to that of journalists, film, TV, radio and stage directors, designers and illustrators, and photographers, who showed average/median ratios between 1.19 and 1.17.
Chart 3 plots the median and average incomes for the nine groups of arts professionals. Journalists and film, TV, radio and stage directors are clearly in the top end, and actors and dancers, musicians and visual artists at the bottom. The median income for all arts professionals in 2001 was about $32,450 and their average income $39,450. The group closest to these overall averages were authors and related professionals.
Private music teachers, who are no longer in the arts professional group as defined by the Census (because they are now classified with other educationalists), had marginally higher median and average incomes than the visual arts and crafts professionals. Table 5 shows the detail.
If we had plotted the four main categories of musicians on Chart 3, music directors would have come third after film, TV, radio and stage directors. Composers would lag behind both designers and illustrators on median incomes, but would be virtually on a par with designers and illustrators on average incomes because of their relatively unequal income distribution. Singers and instrumental musicians had median and average incomes slightly below those pertaining to total musicians (this has to be the case since music directors and composers had higher incomes and therefore brought the average for total musicians up).
Knowing both the total numbers of arts professionals and having estimated their average incomes, we are in a position to estimate the total gross incomes of each group (summarised in Chart 4, which shows the four largest groups of arts professionals and the other groups combined). The nine groups of arts professions (98,000 persons) earned an estimated $3.87 billion, the largest groups being designers and illustrators with 32,200 members ($1.3 billion), and journalists and related professions with 16,500 members ($881 million).
These two groups accounted for just under half the total number of arts professionals, but an estimated 56.5% of total arts professional incomes. They were followed way down the line by about 6,850 film, TV, radio and stage directors earning an estimated $358 million, 9,000 musicians and related pro-fessions ($269 million), about 6,850 photographers ($232 million) and 9,800 visual art and craft professionals ($230 million). The smallest groups in terms of both numbers and total income were actors and dancers numbering about 6,150 ($192 million), about 4,100 authors ($163 million), and less than 2,600 media presenters ($100 million).
Private music teachers, despite their low incomes, had a total income of about $188 million in 2001, almost exactly the same amount as the total number of instrumental musicians. However, this amount took only 6,625 instrumental musicians to earn compared with 8,445 private music teachers.
The income estimates, adding to $3.87 billion for total arts professionals, are compared in Table 5 with the incomes of all cultural occupations as defined by the Australian Bureau of Statistics (and briefly discussed in the last section of this article). The latter group had a total estimated income nudging $10 billion (of which arts professionals accounted for 39%).
The second comparison in Table 5 is with the estimated income of the total employed workforce in Australia, which was $316 billion. This is a crucial finding because one might expect to find a check on this estimate in the Australian national accounts. It would be unreasonable to look for an exact verification, because the Census relies on what Australian residents recalled was their gross weekly income just prior to the Census date and the method of estimating annual income and income distributions is fairly crude. In contrast, the national accounts rely on an array of official statistics and careful statistical checks and balances.
It is pleasing, therefore, that wages and salaries earned by the employed labour force in 2000-01 totalled $305.2 billion and in 2001-02 $319.9 billion, very much in the ballpark of our Census estimate of $316 billion. There are of course other sources of earned income such as property income and commissions, but they would be relatively minor for the active labour force. The total annual estimated household income derived from the 2001 Census income distribution for the adult population was $410 billion, of which the employed labour force accounts for about 78% ($316 billion) and people not in the employed workforce for 22% ($94 billion).
Income trends have varied significantly among arts professionals according to our analysis of Census records compiled between 1986 and 2001 (Table 6). During this period, there was a tendency for median incomes of all arts professionals to fall slightly between 1986 and 1991 (the main exception being the large group of designers and illustrators), stabilise between 1991 and 1996, and then increase significantly between 1996 and 2001.
These income estimates are expressed in September Quarter 2006 prices (Consumer Price Index), which approximate the price level around the 8 August 2006 Census.
Visual artists have been split into two groups in Table 6 because potters and ceramic artists were shown separately in previous years. They have the lowest income levels among the resulting ten groups of arts professionals.
Musicians and related professionals emerge as the only group of arts professionals suffering reductions in real median incomes during all three five-year periods covered in Table 6 (with a compound annual fall of 1.0% in the first period, followed by more moderate annual falls of 0.4% and 0.2% in the two subsequent periods). At the other end of the spectrum, only designers and illustrators enjoyed a consistent rise in incomes throughout the fifteen years. During the last period, 1996-2001, only two groups suffered falls: media presenters (significantly) and musicians (less so).
The range of median incomes of arts professionals widened over the fifteen years. In 1986, the highest median (for film, TV, radio and stage directors) was 138% higher than the lowest median (potters). By 1991, the top group (directors again) had increased the gap over potters to 173%, by 1996 to 182% (when journalists took over from directors as the top group), and by 2001 to 195% higher.
The median income of musicians relative to the median income of all arts professionals fell from 82% in 1986, 79% in 1991 and 78% in 1996, to 71% in 2001. Musicians’ incomes have clearly lagged behind those of almost all other arts professionals. The evidence is summarised in Chart 5, which shows total change in real median incomes between 1986 and 2001.
The median income was significantly higher in 2001 than in 1986 for designers and illustrators, journalists, directors, and actors and dancers. There was little difference in the level of income for visual artists (other than potters) and authors. The main groups suffering losses were photographers (-5%), media presenters and musicians (both -8%) and potters (-12%).
We formed a hypothesis during this period of analysis that the smaller number of musicians included in the Census in previous years were dominated by relatively highly paid orchestral and other full-time musicians, but that a growing popular music industry had added lower-paid members to the Census total. As shown below, this is not right. Furthermore, we are now reminded by the 1991 Census that about three-quarters of singers were popular rather than concert or opera, and have surmised that a similar ratio might exist for instrumental musicians.
Table 7 compares the income distributions of musicians and related workers for the four Census takings between 1986 and 2001, expressed in constant (2006) prices. Because the income classes changed from Census to Census, and were then inflated by the CPI, the material is best viewed graphically using the cumulative income distributions (percentage of people earning more than a given amount). This is done in Chart 6.
The three gridlines on Chart 6 are $20,000, $40,000 and $60,000. Interpolating between the known points on each cumulative income distribution, we find the changes in constant 2006 prices shown in the left-hand table. The proportion of musicians on incomes less than $20,000 increased slightly (from 33% to 36%) between 1986 and 2001, which may be a function of the reductions in median income (using average incomes would have given similar results). However, the proportion of high incomes also increased, and more significantly in relative terms (from 7% to 11% of total musicians). This runs counter to our hypothesis that larger proportions of lower paid popular musicians have joined the full-time workforce. In short, the income distribution has become less rather than more equitable, with the proportion of middle incomes ($20-60,000) falling from 60% to 53% and both extreme income groups increasing their share.
So our hypothesis, such as it was, is refuted. If anything, more musicians were worse off in 2001 than in 1986 as measured by the greater number of musicians and the higher proportion of low incomes. The implication is that one-third of the 6,080 musicians in the 1986 Census were paid less than $20,000 at 2006 values (just over 2,000 persons). In 2001, 36% of 9,000 musicians were in that situation (3,200 persons, or an increase of over 60%).
On the other hand, 7% of the 6,080 musicians in 1986 were paid $60,000 or more (at 2006 values), that is, about 425 persons. In 2001, 11% of 9,000 persons earned such incomes (just under 1,000 individuals).
In 2001, 30.7% of musicians counted in the Census worked full-time, defined as 35 hours or more per week. Composers and music directors were much more likely to work full-time (65% and 64%, respectively). Only 28% of instrumental musicians and 22.5% of singers worked full-time in their profession (Table 8).
Among arts professionals, musicians were the group with the lowest percentage of full-time workers. Over 70% of film, TV, radio and stage directors, designers and illustrators, and journalist and related professionals worked full-time. So did 64% of professional photographers, 58% of authors, 53% of visual artists, 43% of media presenters, and 39% of actors and dancers.
Only 18.6% of private music teachers worked full-time.
The full-time component for musicians increased from 29.4% in 1996 to 30.7% in 2001, though the increase was concentrated on instrumental musicians, while the proportion of full-time workers fell among composers, music directors and singers.
The number of full-time private music teachers in-creased from 17.7% to 18.6% of total music teachers between 1996 and 2001 (Table 9). There is some relationship between number of working hours and incomes, as illustrated by Chart 7. The chart shows four different symbols and two lines illustrating the correlation between hours worked and the proportion working full-time. Two of the symbols relate to only one observation: private music teachers (green triangle) and total arts professionals (orange triangle).
The best correlation applies to the four categories of musicians: two at relatively low levels of income and low working hours (instrumentalists and singers) and two at higher incomes and working hours (composers and music directors). They are shown as blue squares with a blue line indicating the correlation. Very approximately, the relationship suggests that if the proportion of full-time workers increases by 20 percentage points (say, from 20% to 40%), annual median income increases by about $9,300 at 2006 values.
The correlation is much more tenuous between the various groups of arts professionals (red diamonds and a red line showing the correlation, such as it is). Specifically, dots below the line indicate incomes higher than the correlation line ‘explains’; this applies to musicians, actors and dancers, and media presenters. Please note that this is attributable to the weak correlation and there is absolutely no suggestion that these professions are overpaid.
Conversely, the left-hand red dot above the line suggests that visual artists ‘should’ be paid much more to compensate for their relatively long working hours than is indicated by their median income of only $23,500. So, at higher levels of income, ‘should’ photographers (median income $33,900), and designers and illustrators (median income $40,500).
The weak correlation is one indication that we don’t live in a world of simple relationships but in a complex society. Though there is some relationship between the working week and the level of income, it is not strong. The issues obviously differ from profession to profession. For many musicians they would include the multiple activities that many have to undertake, even if their ‘main’ occupation in the week before the Census was performance.
Table 10 shows number of musicians and other arts professionals in each State and Territory in 2001. Table 11 shows percentage distributions.
One appropriate comparison as far as musicians are concerned is with the distribution of the total employed labour force, and with other arts professions. Chart 8 summarises the information. A sketchy State-by-State description follows.
These patterns indicate that many factors come together to determine the geographic distribution of musicians and other arts professionals (and we haven’t even touched the regional detail within States and Territories). Among musicians, music directors and composers seem to gravitate towards the bigger centres, especially Sydney, whereas instrumental musicians in particular are more evenly distributed across the continent. Among other arts professionals, journalists and directors also appear to be drawn towards the bigger population centres where the main employment opportunities are, whereas the opposite is true for visual artists who find havens in places like WA, Tasmania and the Northern Territory.
Employment in culture is a concept developed by the Australian Bureau of Statistics and the Statistics Working Group of the Cultural Ministers Council (SWG) over the past several years. It is based on an occupational and an industry classification.
Several of the tables already discussed refer to total cultural occupations, of which the arts professions that have been our primary interest accounted for 38%. Of the rest, some other occupations are classified as professions, but technical, trade and non-trade occupations are also represented among cultural occupations.
Cultural industries include publishing and printing, broadcasting and film, music and theatre productions, creative arts, libraries, museums, parks and gardens, photographic studios, design, religious organisations and other activities. For musicians and related professions, the most important cultural industry by far is music and theatre productions, which employed 5,500 persons in 2001 (Table 12). Another 770 were employed in creative arts and other cultural industries.
Altogether, 70% of musicians were employed in cultural industries, and 30% elsewhere.
The concept depends heavily on what is considered a cultural industry. In contrast to musicians, only 4% of private music teachers were employed in cul-tural industries in 2001, because education is not part of the definition.
Looking at the wider range of arts professionals, only 40% of the largest group of all, designers and illustrators, were employed in cultural industries. The distinction becomes more useful, though possibly the concept of cultural industries could become outdated in the process, if we introduce the notion of creativity (the subject of a forthcoming article in this knowledge base). Briefly, the concept of creativity as the driver of new economic growth has won prominence during the past ten years. Stuart Cunningham in a recent essay (What Price a Creative Economy? Platform Papers, Currency House, July 2006) has dubbed the following distinction from the 2001 Census the ‘creative trident’ (he also provided the attached statistics):
A similar ‘cultural trident’ may be possible through further study of the 2001 Census. Cunningham specially mentions the role of designers and of other arts professionals in spreading creative activities across the spectrum of industries, and says that “people who are ‘embedded’, i.e. employed in creative occupations outside of specific creative industries, and are spread across all industry divisions, constitute almost 2% of the Australian workforce.” (pp 20-21)
The wider classification of cultural occupations in Table 13 shows that the ratio of people employed in cultural industries and those who according to Cunningham’s concept are ‘embedded’ in other industries varies widely. Some of the variation is due to the definition of what is ‘cultural’, including the exclusion of education from that context.
This may have some justification in the statistical classification of leisure and culture industries developed by the ABS and SWG. However, the development of creativity as a potent driver of economic growth coincided with the time it took to develop the cultural statistical concepts. Key players in that development was the American Richard Florida and the Creative Industries Taskforce set up by the Blair Government in 1998. The latter resulted in a number of governments, including Australia and Queensland, setting up Creative Industries Cluster Strategies with varying emphasis on digital technology versus arts and cultural pursuits.
The time may be ripe to marry the two concepts, concentrating on what is truly cultural creativity and how it can be nurtured and evolved to help the formation of a better cultural and economic base for the future.
Hans Hoegh-Guldberg. Last updated 21 January 2008