VoxEU Column Global crisis Global economy

Causes of the G7 fixed investment doldrums

The post-Crisis G7 economies have suffered weak business investment despite record low interest rates and the favourable financial positions of corporates. Some consider this the ‘new normal’ arising from secular, supply-side forces that have contributed to declining potential growth rates. This column argues that structural factors alone are not sufficient to explain the current weakness in investment rates. There is thus room for positive surprise if companies realise the pent-up investment demand.

Investment has been disappointing in recent years

Growth in mature economies has consistently disappointed in the years following the Global Crisis, and forecasts are regularly being revised downwards – just recently again by the IMF. An important part of the sluggish recovery in mature economies has been weak fixed investment. Total investment relative to GDP in the G7 economies stood at 19.3% in 2013 – a decline of 2.6 percentage points relative to 2007. Part of this weakness in investment can be explained by a fall in residential investment after the bursting of the housing bubble in several G7 countries, and also curtailed public investment spending due to fiscal consolidation efforts.

However, even after stripping out these two components, business investment (i.e. investment in machinery, equipment, transport, structures, and intangible assets) has exhibited an unusually weak recovery. In the second quarter of 2014, G7 private non-residential investment amounted to 12.4% of GDP, compared to the peak of 13.3% in 2008 (Figure 1).[1] This trend has been broad-based across G7 countries, despite differing stages of economic cycles. For example, private non-residential investment rates to GDP were below the pre-Crisis peaks by around 2 percentage points in Italy and Germany, and by 1 percentage point in Japan, France, and the US. Only in Canada and the UK has the gap been smaller, at 0.3 and 0.4 percentage points, respectively.

Figure 1. G7 private non-residential fixed investment

…despite favourable financial market conditions

The recent weakness in G7 investment is particularly surprising given ultra-easy monetary policy conditions, with interest rates across the major currency areas standing at historical lows. As a result, borrowing costs – as measured by bank retail lending rates and corporate bond yields – have declined markedly. At the same time, corporations’ external borrowing needs have declined as internal funds have grown in the years after the Crisis. Operating corporate cash flow, or saving, rose in most G7 countries, notably in the US, Japan and Germany.[2] For example, the saving rate of US non-financial corporations surged to 21% of gross value added, compared to the pre-Crisis level of around 16%. While G7 corporate saving has declined more recently, this to some extent reflects dividend payouts and share buybacks. Even so, net lending as a share of GDP remains positive (Figure 2), as during the period 2002–2005, which was also characterised by low investment spending and high cash accumulation (IMF 2006).[3]

Figure 2. G7 non-financial corporations’ net lending

Source: Haver, US Flow of Funds.

A continuation of a long-term trend?

One important feature of the weakness in investment is that it actually predates the Global Crisis. In fact, G7 investment was on a long-term downward trend well before 2008. For example, at the recent peak of 2007, the aggregate investment rate for G7 countries was 1.2 percentage points lower than in 2000 (Figure 3). By country, investment relative to GDP has exhibited a downward trend since the 1990s in Germany, UK, Japan, and Italy. Investment in the US and France has held up somewhat better, but only Canada has experienced an upward trend over the past two decades, largely reflecting heavy investment in the resource sector.

Partly, this long-run phenomenon reflects a measurement issue, notably lower relative prices of investment goods. Over the past two decades, prices of IT equipment have risen much less than the GDP deflator, while the IT share of total investment (in volume terms) has risen. When correcting for this price effect, the fall in non-residential investment to GDP ratios is much less pronounced (Figure 4). In fact, the ratio of G7 investment in equipment, machinery, and intangible assets relative to GDP has moved upward over the long term, although current levels stand below the trend. However, the real investment rate in non-residential structures still exhibits a long-term downward trend, partly reflecting a decline in public infrastructure investment (McKinsey 2010).

Figure 3. G7 total investment rates

Source: IMF.

Figure 4. G7 real and nominal investment rates (percent of GDP; equipment includes transp. , mach., soft)


Sources: Haver, Eurostat.

The role of structural, supply-side forces

Putting aside cyclical effects and measurement issues, an important factor behind the long-term trend decline in real investment rates in mature economies is related to a decline in growth on the supply side of the economy, reflecting slower labour force growth and the pace of technological innovations. Looking at five-year periods to abstract from the cycle, there is a clear two-way relationship between weaker GDP growth and lower investment to GDP ratios spanning over the past three decades, as well as between weaker investment and labour force growth (Figure 5).

The link between investment rates and labour force growth can be explained in the context of a Solow growth model. According to the Solow model, the capital–output ratio is stable in the long run, with the investment rate being equal to the sum of capital depreciation, growth in the labour force, and technological advances. With the labour force growing more slowly or even declining in mature economies, some economists have proposed that the current investment rates represent a new normal low (Pinto 2014, Summers 2013). However, even taking into account slower growth of the supply side, an increase in investment seems likely.

First, even if we make a conservative assumption about the capital–output ratios (i.e. no further capital deepening), the current investment rates are too low. The rate of GDP growth that is implied by current investment rates is only around 1% for the G7 aggregate. In other words, the current investment rates in the G7 only make sense if long-run GDP growth rates were expected to remain at current depressed levels. This seems unlikely. For example, the OECD estimates potential growth rates for the G7 to be around 1.9% five years from now – roughly twice what is implied by current investment rates. The same relationship used above implies that investment rates would need to rise by an average of more than 3 percentage points to be consistent with a stable capital–output ratio at a potential growth rate of 1.9%.

Second, most economic models predict that slowing labour force growth would be accompanied by some capital deepening, resulting in a higher equilibrium capital–output ratio (given an unchanged saving rate). For example, it is increasingly possible to replace factory workers with robots, and to substitute labour with capital in service industries such as banking (Hamm 2007). Capital deepening should be also facilitated by lower relative prices of investment goods (Karabarbounis 2013). However, the past few years have featured capital shallowing instead.  A long-term downward trend in the US and UK is probably reflecting sustained shifts from manufacturing to the less capital-intensive service sector. But the capital–output ratio and also capital intensity (as measured by capital per person employed) started to decline also in manufacturing-intensive countries such as Japan and Germany (Figure 6). This all suggests an upside potential for capital accumulation and therefore that, over the cycle, investment rates should be higher than just GDP growth and capital depreciation.

Figure 5. G7 real investment rate vs. real GDP growth

Source: Datastream.

Figure 6. G7 capital–output ratio

Source: Datastream.

Structural slowdown, cyclical pickup?

Abstracting from such long-run factors, investment should also benefit from a cyclical pick-up in the economy after many years of sub-par growth, underpinned by an improvement in business confidence across countries (Figure 7), as we argue in a recent research note (Institute of International Finance 2014). Capacity utilisation also has increased, and is standing near long-term averages (Figure 8).

Over the past several years, forecasts for investment recovery repeatedly turned out to have been overly optimistic. While this suggests some potential for disappointment going forward, there is also the potential for a positive surprise if companies realise the pent-up investment demand. Already, business investment rates have picked up since last year in the US, Japan, and the UK, but an improvement is yet to be seen in continental Europe.  Overall, there is some hope that business investment will finally come out of the doldrums – a rare positive feature in a still fairly downbeat economic outlook.

Figure 7. G7 industrial confidence and real investment in equipment

Source: Datastream.

Figure 8. G7 capacity utilisation – manufacturing


Congressional Budget Office (2009), “How Slower Growth in the Labor Force Could Affect the Return on Capital”.

Erber G and H Hagemann (2013), “Growth in Investment Dynamics in Germany after the Global Financial Crisis”, DIW Economic Bulletin.

Hamm I, H Seitz, and M Werding (2007) “Demographic Change in Germany: The Economic and Fiscal Consequences”.

Institute of International Finance (2014), “Business Investment in the G7: Coming out of the Doldrums?”.

International Monetary Fund (2006), “Awash with Cash: Why are Corporate Savings so High?”, World Economic Outlook, Chapter IV.

JPMorgan Chase & Co. (2014), “Are we underinvesting? (With apologies to Piketty)”.

Karabarbounis, L and B Neiman (2013), “The Global Decline of the Labor Share”, NBER Working Paper 19136.

McKinsey (2010) “Farewell to Cheap Capital”.

National Research Council (1986), “Population Growth and Economic Development: Policy Questions”.

Organisation for Economic Co-operation and Development (2007), “Corporate Saving and Investment: Recent Trends and Prospects”, Economic Outlook 82.

Pinto E and S Tevlin (2014), “Perspectives on the Recent Weakness in Investment”, FEDS Notes.

Summers L (2013), “On Secular Stagnation”, Reuters.


[1] UK business investment excludes the Q2 2005 exceptional transfer of nuclear reactors from a public corporation to the central government.

[2] Gross saving (also referred to as operating cash flow in the text) equals undistributed profits (after taxes, interest rates, and dividends) plus capital consumption (depreciation of fixed assets).

[3] The financial account is used for Japan, whereas the capital account is used for other countries. In the capital account, net lending/borrowing equals gross saving (profits after dividends) plus capital transfers (e.g. investment grants), minus gross fixed capital formation and minus net acquisitions of non-financial non-produced assets (such as land and licenses). In the financial account, net lending/borrowing equals acquisition of financial assets. Conceptually, the two indicators are the same, but in practice they deviate due to different calculation methods and sources.

1,890 Reads