Monday, March 16, 2015
We're not taking productivity seriously
Saturday, March 14, 2015
EXPLORING STRUCTURAL CHANGE IN THE AUSTRALIAN ECONOMY
Comview 2015
Big changes in the structure of our economy have been happening before our eyes in recent years. But you well know - and each of your students needs to learn - that there’s nothing new about structural change. It may move faster in some periods than others, but it has been happening continuously in Australia since white settlement. It’s fascinating to trace changes in the structure of our economy, but it’s also illuminating, helping improve our understanding of the economic development process. In what follows I’ll be quoting extensively from the 2014 Australian Industry Report, produce by the chief economist of the federal Department of Industry, Innovation and Science.
What is structural change?
Structural change refers to shifts in the distribution of output, investment and employment across industries or regions. It refers to medium to long-term changes, not short-term changes attributable to the economy’s movement through the business cycle.
Structural change reflects the collective responses of individuals and firms to changing relative prices. Without structural adjustments, economies cannot respond to changes in relative prices, and therefore can’t achieve optimal allocation of resources under new conditions. The decade-long mining boom, for instance, would not have been possible without a rapid reallocation of resources towards that industry and away from other, less profitable industries.
Structural change is a continual process as relative prices are constantly fluctuating and economic resources constantly flow into and out of industries to take advantage of this. To show the economy’s constant state of flux, in the 12 months to February 2013, more than a million workers changed jobs, 9 per cent of the workforce. Of these, about 600,000 changed industry and about 460,000 changed occupation. Similarly, almost 240,000 new businesses entered the economy and about 300,000 businesses left the economy. The net effect of this constant flux is an ever-changing structure of the economy. The creation of new businesses and the decline of less competitive businesses are key to the economy’s long-term growth.
Prices change a lot more quickly than resources are able to move, however. Workers take time to respond to changes in relative wages because it can be difficult to relocate, or because it takes time to reskill and change occupations. Similar barriers effect the timely adjustment of capital and land use. These barriers can lead to unemployment and other inefficiencies such as idle machinery and equipment, which can have costly economic and social consequences. If adjustment occurs as a result of a short-term shock rather than a longer-term trend, the economy can take time to re-adjust and some businesses may not be able to return to their industry as quickly or easily as they left. So structural change is beneficial, but it does have costs. It may have adverse consequences for individuals, particularly those not easily able to move between industries, and so could lead to long-term unemployment. But it may also benefit workers able to take advantage of the new opportunities it offers.
Drivers of structural change
The main drivers (or sources or causes) of structural change are new technology, globalisation, consumer preferences and government policy. We’ll discuss these in turn.
Technology. Technological change is a basic driver of productivity improvement and economic growth. It has played a major part in globalisation, thanks to advances in transport and communications. In turn, globalisation exposes countries to more competition, which leads to a greater number of technological advances.
Technological change includes the development of new products, improvements to existing products and improvements to the processes by which products - services as well as goods - are produced and delivered to customers. So while much technological advance comes via machines or software, some comes via new ways of organising work in factories and offices.
Closely tied with technological change is the diffusion of knowledge, which determines how quickly and efficiently new technology emerges and can be adapted by industry for various purposes. This, in turn, can result in changes in the skills required of workers. Both the pace at which new technology can be adapted and the degree to which labour can be shifted between industries can alter the competitiveness and performance of firms. It’s the behaviour of firms that ultimately drives structural changes in the economy.
Technological change has driven structural change across the economy in a number of ways. Developments in computers, for example, have increased the range of occupations that can be automated, while improved communications and transfer of data have allowed transactions to occur over greater distances. These developments have allowed a greater variety of services to be exported without the need for the physical movement of people between countries.
Globalisation. Globalisation refers to the increasing connections between economies and between particular markets within each of those economies. In recent decades this process has been driven by a combination of technological advances - which have reduced transport costs and improved flows of information - and changes in government policies that have reduce barriers to trade, flows of capital and workers’ movements between countries.
In Australia, effective rates of protection of locally-made manufactured goods have fallen from 35 pc on top of production cost in the late 1960s to less than 5 pc today. Freight and insurance costs for imported goods have fallen from 8 pc to 5 pc since the late 1980s. Over the past two decades, Australia’s international trade (exports plus imports) has increased from 26 pc of GDP to 47 pc. But increasing international linkages also mean our economy is more exposed to international shocks.
Consumer preferences. Changes in consumer preferences also induce changes throughout the economy. As real incomes rise over time, households tend to spend the increase differently to the pattern of their existing spending. The increase goes disproportionately to luxury goods (including leisure activities, travel and fine dining) or to superior goods (such as health, education and housing) and “positional goods” (goods and services that demonstrate your higher social status). The biggest change in consumer preferences, therefore, has been slower growth in spending on goods and faster growth in spending on services. Since the beginning of the 1960s, the proportion of household disposable income spent on goods has fallen from 49 pc to 27pc, while the proportion spent on services has risen from 37 pc to 63 pc (with household saving falling from 14 pc to 10 pc).
As well, the ageing of the population is changing the composition of consumer spending. Over the past four decades, the median age of the population has risen from 27 to 37 years.
Government policy. All government policy potentially affects the sectoral composition of the economy, including by its effect on relative prices. Policy changes may have the effect of driving structural change or influencing the way or extent to which it occurs - eg the state governments’ response to the advent of Uber. Sometimes policy is motivated by a desire to prevent structural change.
Changes in the composition of government spending will affect the allocation of resources in the economy and thus its industry structure. Industry policy is obviously intended to affect the structure of industry. Even a policy to put a price on carbon is intended to bring about structural change, with more growth in renewable-energy industries and contraction in fossil-fuel industries.
The move from centralised wage-fixing to bargaining at the enterprise level has allowed relative wages to adjust more efficiently across industries and regions and given workers a stronger price signal of the relative demand for skills. It has also helped to prevent the wage inflation that has limited the gains from commodity price booms in the past.
The long-term pattern of structural change
The pattern of structural change in Australia over the past century or more is similar to that experienced in other developed countries. Stage one: pre-industrial economies are dominated by agriculture.
Stage two: productivity improvements due to technological advances free up resources, particularly labour, that were producing food to work in other industries, as less labour is required to feed the population. This results in the long-term decline of agriculture and the rise of manufacturing.
Stage three: as economies further modernise, manufacturing declines and services emerge as the dominant sector. This is typically a result of a more than proportionate change in demand from goods to services as incomes increase.
In Australia, the share of output from agriculture fell from over a third in the 19th century to just 3 pc in the 2000s. Similarly, the share of manufacturing output and employment fell by more than half between the 1960s and 2000s - output falling from 26 pc to 12 pc and employment falling from 26 pc to 11 pc. While services already accounted for about half of output in the 19th century, their share had grown to 59 pc by the 1960s and to 78 pc by the 2000s.
Note that a decline in a sector’s share of total output or employment may or may not imply an absolute decline in its output or employment. Despite the dramatic decline in agriculture’s share of total output over the past century, the quantity or volume of its output has continued to grow, by a factor of six. Never in human history have we feed so many mouths with so few farm hands. So agriculture’s share of the economy has declined not because it has been contracting, but because other industries have grown faster than it has.
Recent structural change in Australia
Mining. The resources boom, which lasted about a decade from the early noughties, had big effects on the business cycle, but nonetheless constituted a major and lasting change in the structure of our economy. This is because the boom in the prices we received for our exports of coal and iron ore prompted a huge expansion in our capacity to produce minerals and energy, including natural gas. Over the decade to 2014, mining’s share of total output doubled to 10 pc - probably the highest it’s been since the days of the gold rush.
You might expect an industry’s share of total output to be similar to its share of total employment, but this isn’t always the case. It depends on the extent to which its production is labour-intensive or capital-intensive. Mining is our most capital-intensive industry, so although its share of total output has doubled to 10 pc, its share of total employment has increased only from 1 pc to a little more than 2 pc.
Our other capital-intensive sectors are the other goods-producing industries, agriculture and manufacturing. Continuous advances in labour-saving technology have made them progressively more capital-intensive. It follows that the services sector accounts for most of our labour-intensive industries.
Manufacturing. Manufacturing’s shares of total output and total employment have been decline since the mid-1960s, with the share of employment falling faster than the output share because of computerisation. Note that, despite an absolute fall in its employment, the absolute quantity of manufacturing output continued growing until 2008. This, of course, is evidence of the industry’s ever-improving labour productivity. From 2008 on, the exceptionally high exchange rate caused by the mining boom greatly reduced the industry’s international price competitiveness, causing its output to fall in absolute terms and its declining shares of output and employment to accelerate. It remains to be seen to what extent the industry recovers now the exchange rate has returned to more comfortable levels.
Manufacturing employees are generally less geographically mobile than in other industries. Even so, manufacturing in regional areas has demonstrated considerable capacity to adjust in the face of the recent decline in employment. Employment outcomes for automotive workers have been better than feared. More broadly in manufacturing (and other declining industries), workers were mainly able to transition smoothly to new employment, as unemployment among these workers was not significantly higher than across all industries. Over the five years to 2011, the largest proportion of workers who moved into mining came from manufacturing - about 18,500 of them.
Despite impressions to the contrary, manufacturing remains an important part of the Australian economy, producing about $100 billion-worth of output each year and so making it the sixth biggest industry. It still employs more than 930,000 workers, making it our fourth largest employing industry. And not all parts of manufacturing are declining. The food, beverage and tobacco industry increased its employment by nearly 50,000 over the decade to 2012, making it the largest employer in the sector. Job losses have been greatest in textiles, clothing and footwear, and machinery and equipment manufacturing.
The relative decline of manufacturing can be attributed to various factors as well as reductions in import protection: increased competition from low-cost countries such as China and India and a shift in domestic consumer preferences towards services.
Manufactures have accounted for about 80 pc of the value of Australia’s total imports over the past four decades, a sign of our lack of comparative advantage in this sector. About half these imports come from other OECD countries, with China accounting for about a quarter.
Services. The Australian economy is transitioning to a knowledge-based economy. This implies a smaller role for primary and secondary sectors and an even greater role for services in the economy. As well, a richer and older population is demanding more health care, more tourism and more education.
This shift to services is associated with changes in the skill and education levels of employees. Areas of employment growth are dominated by tertiary qualified workers, while employment losses are concentrated among those without post-school qualifications. The long-term structural shift towards service industries has resulted in a vast increase in highly skilled occupations, such as professionals (accountants, lawyers, engineers etc) and managers. This suggests that the structural shift towards services has necessitated a transition to an increasingly highly-skilled workforce. The resulting increase in demand for highly skilled workers can also be traced back to technological change. The rise of computer technology, for example, has led to an increase in demand for workers with the skills to use and maintain them. This is known as skill-biased technological change.
Structural change by state
Just as the industrial structure of the Australian economy is always changing, so its spatial structure across the continent is always changing. Increases in population cause increases in economic activity, but increases in activity in particular industries also attract higher population. There is much internal migration between Australia’s six states.
The oldest states - NSW and Victoria - have the biggest populations and shares of the national economy, but the younger states - particularly Queensland and Western Australia - have had the fastest rates of population and economic growth in recent decades, meaning their shares of the national population and economy are growing at the expense of the other states. As we were reminded by the resources boom, Western Australia and Queensland also enjoy a greater natural endowment of minerals and energy. This has left South Australia and Tasmania’s shares of the national population and economy small and getting smaller. They don’t have a lot of minerals, are suffering from the decline of manufacturing and aren’t at the centre of the growing business services sector.
I now turn from quoting the Australian Industry Report to quoting heavily from an article about The Economic Performance of the States in the March quarter, 2015, issue of the Reserve Bank’s Bulletin.
Over the decade to 2013-14, real gross state product has grown at an annual rate of almost 5 pc in WA and 3.5 pc in Queensland, compared with 2.5 pc or less in the other states. This has caused NSW’s share of Australian GDP to fall by 4 pc points to 31 pc. Victoria’s share has fallen by 2 pc points to 22 pc, while Queensland’s share has increased by 1 pc point to 19 pc and WA’s share has increased by 6 pc points to 17 pc. SA’s and Tasmania’s shares have fallen a little to 6 pc and 2 pc respectively.
You would expect the states’ population shares to be similar to their shares of the national economy. This is generally true, but with two notable exceptions. Victoria has 25 pc of the population (and of national employment) but only 22 pc of the economy (which may imply that it has a relatively high proportion of low-paid service sector jobs). And WA’s heavy dependence on capital-intensive mining means it has a 17 pc share of the economy, but just 11 pc of the national population.
For the most part, each state’s industrial structure is not very different from the national average, measured by output. But there are some notable divergences. NSW’s greatest strength is in business services, which account for 30 pc of its total output. Victoria’s manufacturing sector accounts for 7 pc of its total output, little different from the national average of 6 pc. Its greatest strength turns out to be, like NSW’s, business services, accounting for 27 pc of its output, compared to an average of 16 pc for the other four states. Queensland’s industrial structure is surprisingly similar to the national average, apart from a larger mining sector. But WA’s mining sector really dominates its economy, accounting for an amazing 30 pc of its output, compared with an average of about 2 pc for the other four states. SA and Tasmania are notable for their well above-average dependence on agriculture.
Good graphs from the Australian Industry Report for 2014
Chart 2.11 Output share by sector (1993-94 - 2013-14) report page 91 (PDF page 105)
Chart 2.16 Employment share by sector (1993-94 - 2013-14) report page 97 (PDF page 111)
Chart 2.17 Change in employment (2003-04 to 2013-14) report page 98 (PDF page 112)
Chart 2.23 Employment growth by occupation (2004 - 2014) report page 107 (PDF page 121)
Table A3: Industry Share of State Production
Good tables from RBA Bulletin, March quarter, 2015, article, The Economic Performance of the States
http://www.rba.gov.au/publications/bulletin/2015/mar/pdf/bu-0315-2.pdf
Appendix A
Table A1: Relative Size of States, 2013-14