The productivity perfect storm

In today’s RBA Bulletin, Patrick D’Arcy and Linus Gustafsson produced a fascinating study of what has caused Australia’s post-millennium productivity slump. Called Productivity Performance and Real Incomes, this thing is a piece of work.

It begins with a chart of the slump:

Then blames a fair slice of it on the mining and utilities sectors:

In the case of the mining industry, the fall in productivity is partly a natural consequence of the rapid run-up in commodity prices, which has increased the profitability of more marginal deposits. Higher commodity prices justify more difficult and costly extraction of previously undeveloped resources, which becomes necessary over time as developed deposits are depleted. The very rapid pick-up in commodity prices has also justified an unprecedented increase in capital investment in the industry. This growth in measured capital inputs has detracted from measured productivity owing to the lag (of some years) between the initial investments, the completion of projects and the utilisation of all the new capacity. In effect, the productivity developments in the mining industry are best characterised as a movement up the industry’s supply curve, rather than an exogenous shift in the supply curve related to some fundamental change in underlying productivity.

And for utilities:

The fall in the level of productivity in the utilities industry is also related to large investments, which have been necessary to deal with some of the fundamental structural challenges facing the industry, but these investments have not necessarily resulted in higher quantities of measured output. Part of the surge in investment over recent years reflects a significant catch-up that has required rapid
growth in utilities’ workforces after a period in the 1990s when investment and employment in the industry were falling.

There has also been additional investment to improve the reliability of supply in the electricity and water industries, which has only made a marginal contribution in terms of additional measured output. One example is recent investment in desalination plants that, with the return to high rainfall in recent years, are not currently being utilised fully, but will provide a source of fresh water in the event of future droughts.

But it does not stop there. The study goes on to note that wider industry productivity growth rates also fell:

Multifactor productivity outcomes in the 2000s were clearly weaker than the period of strong growth in the 1990s. However, the difference between trend growth in the 2000s and the long-run average prior to the 1990s is less marked. For the market sector excluding mining and utilities, the average growth in multifactor productivity of 0.4 per cent in the 2000s is only 0.2 percentage points lower than the average for the market sector in the period 1973/74 to 1993/94. This suggests that it is the 1990s that was the period of exceptional growth.

The study then goes on to examine why the 1990s period was exceptional:

That productivity growth has slowed across a large number of developed economies in the 2000s provides some indication that there may have been a slowing in the pace at which the technological frontier is expanding. Data on productivity growth for members of the Organisation for Economic Co-operation and Development (OECD) indicate a fairly universal slowing in productivity growth in the 2000s compared with the 1990s, with 19 of 25 countries experiencing a slowdown in productivity growth.

It is difficult to be conclusive about what might have driven this common international experience, but it suggests that part of the slowdown may be related to common global factors, such as the pace of technological innovation and adoption.

This obviously contrasts with the popular version of events widely accepted in Australian economic circles that:

…the acceleration and subsequent slowing in productivity growth over the past two decades relates to the gradual waning of the impetus to productivity growth initiated by the economic policy reforms of the 1980s and 1990s (Dolman 2009; Eslake 2011). These reforms, which included tariff reductions, privatisation, liberalisation of financial markets, decentralisation of the labour market and, somewhat later, national competition policies and tax reform, are widely viewed as having contributed to a marked improvement in economic efficiency.

The overall effect of all these reforms was to increase competitive pressures on firms in product markets such that improvements in productivity became an imperative for economic survival, while at the same time increased flexibility in capital and labour markets ensured that economic resources were allocated more efficiently among competing firms. It is difficult to be definitive about the magnitude of the impact of regulatory reforms, as in many cases, for example with tariff cuts, the changes were introduced gradually over an extended period of time, with the impact on productivity occurring only with a lag.

While some analysts have argued that these reforms should have permanently lifted the growth rate of productivity relative to the unobserved counterfactual, the experience of the past two decades suggests that the effect on productivity growth may have been temporary. Productivity growth appears to have been higher during a ‘catch-up’ period when reorganisation in response to the reforms drove improvements in economic efficiency allowing the economy to move closer to the production frontier.

To this we could obviously add the clear consolidation that we have seen across all industry sectors since 1990s, taking away much of the competitive impulse to find efficiencies.

This study is really worth reading and is quite enlightening but it also has two shortcomings. Take a look at this analysis from the New Zealand Treasury early last year:

New Zealand’s growth in the period leading up to the GFC (2002 to 2008) was associated with rapid credit expansion, fast growth in consumption, high external borrowing, low private saving, a house and farm price boom, high government tax revenues and spending, and static tradable sector growth. Growth on this basis was unsustainable and had several negative consequences including rapid growth of private-sector debt and a high NFL-to-GDP ratio [see below chart].

There are important connections among these factors and the disappointing growth outcomes…

Easy access to credit (particularly for property financing) was associated with rapidly escalating farm and house prices.

The property boom in turn led households to feel wealthier and increase their consumption. Strong growth in nominal GDP buoyed tax revenues and encouraged often ill-judged growth in government spending.

High growth rates of private and public consumption created inflationary pressure that led the Reserve Bank to raise the official cash rate. This, in turn, pushed up the exchange rate.

This standard monetary policy response led to unbalanced growth: reduced exports and increased imports, with a rise in the proportion of domestic resources going to non-tradable production and a reduction in the proportion flowing to tradable production.

It is likely that the reduction in export competitiveness has, in turn, slowed productivity growth. There is evidence that a lower exchange rate creates opportunities for high-productivity firms to grow, achieve scale through exporting, and contribute a range of spill-over benefits to other firms. In contrast, what happened is that non-tradable industries such as construction and property and business services expanded rapidly and attracted labour and other resources at the cost of their availability to export industries such as agriculture and food processing…

Bingo. The same process transpired in Australia but was coupled with a commodities boom that was also drove up the exchange rate. I see no reason why the same effect of displacing tradeables wouldn’t have transpired in Australia, but on steroids.

Put all of these factors together – increased concentration with lowered competition, a fading tech revolution, heavy but not efficiency boosting infrastructure renewal, falling productivity related reform, less efficient mines, consumption displacing tradeables lowering competition further – and you have pressure upon both labour and capital efficiency.

Is it any wonder we got no damn productivity?

bu-0612-3