China: workshop of the world

Global savings glut?

Michael Roberts reviews Matthew Klein and Michael Pettis’s 'Trade wars are class wars'

Klein is the economics commentator at Barron’s, who has previously written for the Financial Times, Bloomberg View and The Economist, while Michael Pettis is professor of finance at Peking University’s Guanghua School of Management and a senior fellow at the Carnegie Endowment for International Peace.

This book has a provocative title, but it is apposite, given the growing global rivalry between the US and China, with the implementation of trade-tariff and technology war. The US is trying to curb and reverse the rising share of trade and hi-tech production that China has been achieving and using to widen its influence globally - at the expense of an ageing and relatively declining US hegemony.

The subtitle of the book is ‘How inequality distorts the global economy and threatens international peace’. Klein and Pettis argue that the origins of today’s trade wars emerge from decisions made by politicians and business leaders in China, Europe and the United States over the past 30 years. Across the world, the rich have prospered, while workers can no longer afford to buy what they produce, have lost their jobs and have been forced into higher levels of debt. Rising inequality has weakened aggregate demand; and a global “savings glut” generated by countries like Germany and China is creating huge global imbalances in demand and supply that threaten economic crises, increased protectionist rivalry and international peace.

The essence of the problem for Klein and Pettis is “the greater eagerness of producers to sell than of consumers to buy”. According to them, this is at the heart of the imperialist rivalry globally. The authors revert openly and clearly to the thesis of John Hobson, the anti-Semitic, social-reformist writer and economist of the early 20th century. They update the Hobsonian thesis for the 21st century. As Pettis puts it,

Our argument is fairly straightforward: trade cost and trade conflict in the modern era don’t reflect differences in the cost of production; what they reflect is a difference in savings imbalances - primarily driven by the distortions in the distribution of income. We argue that the reason we have trade wars is because we have persistent imbalances, and the reason we have persistent trade imbalances is because, around the world, income is distributed in such a way that workers and middle class households cannot consume enough of what they produce.1

Thus, we have a straightforward underconsumption theory of crises, as presented by Hobson. What is added by the authors is the concept of a “global savings glut” - or the reciprocal of a lack of consumption - which generates “global imbalances” between those countries running systematic trade and income surpluses (China, Germany) and those like the US, running chronic deficits. This imbalance of consumption and saving between the major economic powers is the essential cause of future crises and even wars, according to the authors.

What is missing from this analysis is exactly the same as what is missing from all underconsumption theories: namely, capitalist investment. Consumption is not the only category of “aggregate demand”: there is also investment demand by capitalists. Indeed, Marx argued that this was the most important factor in driving growth of production in a capitalist economy - and even Keynes sometimes agreed. I have previously shown that it is capitalist investment that is the ‘swing factor’ in booms and slumps - a fall in investment leads capitalist economies into slumps.2 Consumption is a lagging factor, and indeed changes in consumption are small during the cycle of boom and slump, compared to investment.

Moreover, using International Monetary Fund/World Bank data, if we look at investment rates (as measured by total investment compared to gross domestic product), we find that in the last 10 years total investment (government, housing and business) to GDP in the major economies has been weak: indeed in 2019 total investment to GDP was still lower than in 2007. In other words, even the low real GDP growth rate in the major economies in the last 10 years has not been matched by total investment growth. And, if you strip out government and housing, business investment has performed even worse.

The national savings ratio of the advanced capitalist economies in 2019 is no higher than in 2007, while the investment ratio has fallen 7%. There has been an investment dearth, not a savings glut.3 In my view, this is not the result of a lack of aggregate demand caused by rising inequality and the inability of workers to buy back their own production. It is the result of the declining profitability of capital in the major capitalist economies, forcing companies to look overseas to invest, where profitability is higher (the investment ratio in emerging economies is up 10% in the last 10 years - something Klein and Pettis do not note). As usual with Keynesian and post-Keynesian analyses, the movement of profit and profitability is ignored.


Klein and Pettis like to refer to the work of Atif Mian and Amir Sufi, who emphasise rising inequality from the 1980s - a shift in income from the poorer to the top 1%, leading to a rise in household debt and a “savings glut”. But the latter do not explain why there was rising inequality and they ignore the rise in corporate debt, which is surely more relevant to capital accumulation and the capitalist economy.

Household debt rose because of mortgage lending at cheaper rates, but, in my view, that was the result of a change in the nature of capitalist accumulation from the 1980s, not the cause.4 Actually, in The saving glut of the rich and the rise in household debt, Mian and Sufi hint at this. They note that the rise in inequality from the early 1980s “reflected shifts in technology and globalisation that began in the 1980s”. Exactly. What happened in the early 1980s? The profitability of productive capital had reached a new low in most major capitalist economies - the evidence for this is overwhelming.5

If we are measuring ‘aggregate demand’ by consumption globally, there has been no decline: on the contrary, household consumption as a share of GDP rose to new highs in the major economies. What ended this speculative credit boom was the downturn in the profitability of capital from the end of the 1990s, leading to the mild ‘hi-tech’ bubble burst of 2001 and eventually to the financial crash and great recession of 2008. A “savings glut” is really one side of an ‘investment dearth’. Low profitability in productive assets became a debt-fuelled speculative bubble in fictitious assets.

Crises are not the result of an “indebted demand” deficit, but are caused by a ‘profitability deficit’. The “class war”. which Klein and Pettis argue is the cause of trade wars, is related to the exploitation of labour by capital for higher profitability, not a lack of domestic consumption caused by low wages.

Klein and Pettis follow John Hobson in his argument that imperialism (or trade wars for our authors) was the result of capital being forced to seek new markets overseas because of the lack of consumption demand at home. According to Pettis,

It’s interesting to go back to Hobson. He argued that the reason England and other European countries exported capital abroad was not military adventurism, but income inequality. You had incredibly high savings, because much of the income was concentrated among the wealthy, and so England had to export those excess savings and the accompanying excess production. Imperialism enabled it to lock in markets for both of those exports. Hobson’s prescription was that increasing the wages of English workers such that they’re able to consume what they produce would make imperialism unnecessary - and this is where I see the connection to today.6

This is what Hobson reckoned for the late 19th century, when the UK was the leading imperialist power. But the evidence does not back this up. The great economist, J Arthur Lewis, summed up the driver behind Britain’s imperialist ambitions in the late 19th century:

In the low level of profits in the last quarter of the century we have an explanation which is powerful enough to explain the retardation of industrial growth in the 1880s and 1890s … we have here also, in low domestic profits, the solution to the great mystery of British foreign investment: namely why Britain poured so much capital overseas … home industry was so unprofitable in the 1880s through the squeeze on profits between wages and prices.

Lewis shows that during the long depression nominal wages fell, but, as prices fell more, real wages stayed up at the expense of profits.7

As the Marxist economist of the 1920s, Henryk Grossman, said of Hobson’s thesis, “It is not enough to account for capital export in terms of the lack of profitable investment opportunities at home, as the liberal economist and pioneering critic of imperialism, John Hobson, put it”.8 He went on to ask why, in that case,

... are profitable investments not to be found at home? ... The fact of capital export is as old as modern capitalism itself. The scientific task consists in explaining this fact, hence in demonstrating the role it plays in the mechanism of capitalist production.

It is the race for higher rates of profit that is the motive power of world capitalism. Foreign trade can yield a surplus profit for the advanced country.

From about the 1980s onwards, the rate of profit in the major economies reached new lows, so the leading capitalist states again looked to counteract Marx’s law through renewed capital flows into countries which had massive potential reserves of labour that would be submissive and accept ‘super-exploiting’ wages. World trade barriers were lowered, restrictions on cross-border capital flows were reduced and multinational corporations moved capital at will within their corporate accounts. This explains the policies of the major imperialist states at home (an intensified attack on the working class) and abroad (a drive to transform foreign states into tributaries).

A recent paper by two economists at the US Federal Reserve, Joseph Gruber and Steven Kamin, shows a widening gap between corporate savings (or profits) and corporate investment in most of the major economies.9 But Gruber and Kamin demonstrate that this was because rates of corporate investment “had fallen below levels that would have been predicted by models estimated in earlier years”. With the exception of Japan, since 1998 corporate savings to GDP have been broadly flat. So the gap between savings and investment cannot have been caused by rising savings. There is not too much profit (surplus savings), but too little investment. The capitalist sector has reduced its investment relative to GDP since the late 1990s and particularly after the end of the great recession.

As profitability fell, investment declined and growth had to be boosted by an expansion of fictitious capital (credit or debt) to drive consumption and unproductive financial and property speculation. The reason for the great recession and the subsequent weak recovery was not a lack of consumption or a savings “glut”, but a collapse in investment.

Trade war

Klein and Pettis go on to argue that policies of investment for export by countries like China and Germany create “global imbalances” that encourage dangerous trade wars. According to the authors, some capitalist economies are “saving” too much: ie, not investing at home enough to use up savings and so export abroad, thus running up big trade surpluses. Others are forced to absorb these surpluses with excessive consumption and run large current-account deficits. And so we have trade wars, as governments like Trump’s try to reverse this trend.

Klein and Pettis are saying that these imbalances are caused by the decisions of governments like China and Germany that seek to suppress wages and consumption (the “class war”) in order to boost investment and export surplus savings. As Adam Tooze, the radical historian, put it, the authors “divide the world into the surplus generators and the deficit countries. The strong causal claim is that imbalances are largely the result of social-structural change in the surplus countries.”10

Pettis admits that China’s investment-driven and export-led policy was “not necessarily a bad thing”:

After five decades of anti-Japanese war, civil war and Maoism, China was hugely underinvested. By constraining the ability of households to consume a significant share of what they produced, the government effectively forced up the savings rate and channelled all of those savings into a massive investment programme. China had the highest investment growth rate in history, and the highest investment share of GDP in history. As a developing country that was significantly underinvested, this was exactly what the doctor ordered.11

Following a growth model for developing countries that concentrates on investment (put forward by Alexander Gerschenkron for US development in the 19th century), the authors argue: “It’s a very successful growth model, but, once you reach the point at which you can no longer increase investment at a great pace, you are forced to make institutional reforms.” However, Klein and Pettis reckon:

The problem emerged when the Chinese economy could no longer absorb new investment productively … Once China reached that point, consumption was too low to drive growth, and it entered into a state of excess production.

This is surely nonsense. It is just not true that household consumption in China is being repressed. Actually, personal consumption in China has been increasing much faster than fixed investment in recent years, even if it is starting at a lower base. Consumption rose 9% last year, much faster than GDP.12

Pettis’s and Klein’s own empirical analysis13 reveals that there has been a rise in consumption as a share of GDP in China in the last 10 years - even without recognising that this is a probable underestimate of the size of household consumption in the stats (which exclude many public services or the ‘social wage’).

Class war

But, to repeat, it is not a lack of consumption that is key to the “class war”: it is profitability. In a capitalist economy, companies compete with each other to raise profitability through the introduction of new technologies. There is an inherent contradiction under capitalist production between a tendency for falling profitability of capital alongside rising productivity of labour. As capitalists try to raise the productivity of labour by shedding labour through technology - and so lowering labour costs and increasing profits and market share - the overall profitability of investment and production begins to fall. Then investment collapses and productivity stagnates. I would argue that the recent drop in the investment share of GDP in China is due to the falling profitability of the capitalist sector of the Chinese economy.

Klein and Pettis seem to be on stronger ground in their second case study of an excess-saving surplus economy: Germany. They argue that German politicians after the unification in the 1990s did not want to subsidise the real incomes of those from the east, as it would damage profitability. So they introduced (under a social democrat government) the Hartz reforms, which successfully repressed real wages and boosted profits. This is something I have previously noted myself,14 and Klein and Pettis confirm that story.

But surely here we are also validating the real class war involved: namely between capital and labour over profits and wages; not, as Klein and Pettis see it, the “class war” through restricted consumption. Their idea is:

... the excess savings of Germany and several other smaller European countries (such as the Netherlands) were offset by unsustainable credit and spending booms in countries such as Greece, Ireland and, above all, Spain. The global financial crisis ended that. Since then the entire euro zone has moved into a globally destabilising current-account surplus, in a damaging attempt to turn the second-largest economy in the world into a bigger Germany in the midst of a global savings glut.15

But any proper analysis of the euro imbalances will find that it was not a result of Germany needing to export its “excess savings”, but the result of Germany’s more superior technology and productivity,16 enabling it to expand exports throughout the European Union at the expense of its other weaker member-states. In my view, global imbalances in trade and capital are the result of the higher productivity and technology base of the major companies in the ‘winning’ economies, leading to a transfer of profits to the strong from the weak. It is not a transfer of excess savings to excess consumption across borders, but the transfer of value and surplus value from the weaker capitalist economies to the stronger. Indeed, that is precisely the nature of imperialism: the unequal exchange of value, not a savings-consumption imbalance.

To show this in relation to the Klein-Pettis thesis, I decided to take closer look at savings-investment imbalances. Over the last 30 years, China’s savings rate rose 25.8%, but its investment rate rose more - by 26.8%. So no savings glut there - at least over the long term. Indeed, in the global boom period of the 1990s, China’s investment rate rose much faster than its savings rate. There were no large surpluses on the current account. Only in the short period of 2002-07 did China run a large net savings surplus. In my view, this was because investment and production were switched sharply into exports, as China took advantage of its cost superiority on its entry into the WTO.

In the case of Germany, in the boom decade of the 1990s the savings rate fell, along with the investment rate, as West Germany absorbed the East. Indeed, the investment rate was higher in the 1990s and Germany ran current-account deficits. The big rise in net savings took place after 2002, with the start of the euro and the Hartz reforms. Germany’s investment rate rose, but not as much as the saving rate, as Germany ran big surpluses with other euro zone countries.

In the case of the US, between 2002 and 2018 the savings rate actually rose by 1%. It was the investment rate that fell by 3.2%. Indeed, in the period since the 2008 great recession, the US savings rate has risen 21.3%, while that of investment has fallen 0.5%. Again, the so-called “savings glut” thesis has only some purchase in the short period of 2002-07. Then Germany and China savings rates rose 25%-30%, while the US rate fell 4.2%. But otherwise in the 21st century, the “savings glut” imbalance is a myth.

China’s net savings rate (current account surplus as a percentage of GDP) is now no higher than in the late 1990s. It is true that Germany’s has risen significantly, but the US savings deficit is no higher than at the beginning of the century - so it is no victim of these excess-savings economies. The Klein-Pettis thesis is limited in validity to a short time period and is really old hat now.

Klein and Pettis argue:

The rest of the world’s unwillingness to spend - which in turn was attributable to the class wars in the major surplus economies and desire for self-insurance after the Asian crisis - was the underlying cause of both America’s debt bubble and America’s deindustrialisation.17

But this is historically inaccurate. Since the 1970s, the US had been losing market share in manufacturing and trade and running current account deficits, not just after the Asian crisis. The cause of this decline is down to the relative weakness of US productivity growth, not Asian excess savings. Moreover, US manufacturing companies had shifted their production abroad during the 1980s.

And yet Klein and Pettis want to position the US as a victim of Asian and German economic policy. As Adam Tooze says,

One could read this as an apology for the protectionist turn of American politics in recent decades. You are showing that there is, in fact, a hidden macroeconomic rationale to the desire of American policymakers to ward off other people’s imbalances.18

But, rather than a victim of excess savings economies, the US, as the hegemonic imperialist power, gains extra value from trade and capital flows; tribute from mainly the peripheral economies of the global south (including China); and even to some extent from the likes of Germany. The US is not the victim, and China and Germany are not perpetrators of crises. Instead, the victim is labour everywhere and the perpetrator is capital everywhere. Both American and German workers are being exploited by capital and that is the basis of the class war; meanwhile, the distribution and sharing by capital of surplus value is the basis of the trade war.

Klein and Pettis argue that “America doesn’t control its current account; it doesn’t control its net overall macroeconomic savings balance.” That is true. But it does not need to. On the contrary, as an increasingly rentier economy, it can extract ‘rent’ or surplus value from other, more productive economies through both its external current and capital accounts. And it can do so better than, say, the UK, because it is still the hegemonic power that controls the international reserve currency, the dollar; and it has the financial firepower and military might. It is the Roman empire of the 21st century - in its declining stage, but not yet collapsed.

The Klein-Pettis thesis leads to the conclusion that wages are too high for capital:

In the current environment, the argument against increasing wages is too strong: higher wages reduce competitiveness and cause the benefits of higher wages to flow abroad. If you pay your workers more, consumers in your country will consume from abroad, because prices have to rise. If you believe that the problem is a sort of massive beggar-thy-neighbour problem - in which every country improves its relative position by putting downward pressure on wages, either directly like Germany did under the Hartz reforms, or indirectly through weak currencies and subsidies - then it’s very difficult to raise wages.19

Yes, but that is capitalism: “In fact, you get a situation in which each country benefits by lowering wages,” they say. Exactly, because this boosts profits.


Of course, Klein and Pettis argue that low wages cause crises, given their underconsumption theory. If so, what is the answer to low wages? Pettis states: “In order to address the problem of wages, we have to prevent the unfettered flow of capital. We need some sort of protection. But, rather than trade protection, I would argue that we need to impede capital flows.”20 And the US needs to invest more.

So we must try and make US capitalists invest more by restricting the flow of foreign savings into the economy with capital controls. But will US companies raise investment, while profitability remains low? Of course, for our authors profitability is irrelevant: what matters is reducing ‘excess consumption’ in the case of the US.

For Klein and Pettis, this is the solution to crises. As Klein puts it in explaining where Lenin’s theory of imperialism differs from Hobson’s,

Lenin’s understanding of Hobson was that capitalism inevitably leads to imperialism, which generates conflict among the capitalist powers. But that wasn’t Hobson’s actual argument. He argued that there are problems in the distribution of income and purchasing power within the major European capitalist countries, and that this explains imperialism. That’s an important difference. Hobson’s interpretation was that there are middle courses between overthrowing the entire system and tolerating exploitative international relationships, and we agree. We don’t argue that we’re in an inevitable crisis of capitalism, but rather that the problems we face can be solved using the kinds of redistributional tools that policymakers have used in the past.21

For Klein and Pettis, there is nothing wrong with the capitalist mode of production and investment for profit: it is just the imbalances of savings and consumption they generate that is the problem. Boost wages and reduce inequality and all will be well, as the global imbalances disappear and aggregate demand rises. They declare:

Trade war is often presented as a war between countries. It is not: it is a conflict mainly between bankers and owners of financial assets on one side and ordinary households on the other - between the very rich and everyone else.22

For them, the class war is between “bankers” and “households”, not capital and labour. And the coming imperialist trade wars are between excess savers and excess consumers, not between rival imperialist powers over the share of profits extracted from labour globally.

Which is the more accurate explanation of class and trade wars: that of Klein-Pettis and Hobson or the Marxist one?

Michael Roberts

Michael Roberts blogs at

  1. phenomenalworld.org/interviews/trade-wars-are-class-wars.↩︎

  2. See thenextrecession.wordpress.com/2016/02/19/investment-investment-investment.↩︎

  3. See bis.org/review/r161109d.pdf.↩︎

  4. thenextrecession.wordpress.com/2014/03/11/is-inequality-the-cause-of-capitalist-crises.↩︎

  5. See World in crisis (Chicago 2018),co-edited by Guglielmo Carchedi and myself.↩︎

  6. phenomenalworld.org/interviews/trade-wars-are-class-wars.↩︎

  7. See my book, The long depression (Chicago 2016).↩︎

  8. marxists.org/archive/grossman/1929/breakdown/ch03.htm.↩︎

  9. See thenextrecession.wordpress.com/2015/11/18/savings-glut-or-investment-dearth/gruber-corporate-profits-and-saving.↩︎

  10. phenomenalworld.org/interviews/trade-wars-are-class-wars.↩︎

  11. Ibid.↩︎

  12. xinhuanet.com/english/2019-10/29/c_138512771.htm.↩︎

  13. ft.com/content/f3ee37e0-b086-11ea-a4b6-31f1eedf762e.↩︎

  14. thenextrecession.wordpress.com/2013/09/22/german-capitalism-a-success-story.↩︎

  15. ft.com/content/f3ee37e0-b086-11ea-a4b6-31f1eedf762e.↩︎

  16. thenextrecession.wordpress.com/2019/01/01/20-years-of-the-euro-part-one-has-it-been-a-success.↩︎

  17. ft.com/content/f3ee37e0-b086-11ea-a4b6-31f1eedf762e.↩︎

  18. phenomenalworld.org/interviews/trade-wars-are-class-wars.↩︎

  19. ft.com/content/f3ee37e0-b086-11ea-a4b6-31f1eedf762e.↩︎

  20. phenomenalworld.org/interviews/trade-wars-are-class-wars.↩︎

  21. Ibid.↩︎

  22. ft.com/content/f3ee37e0-b086-11ea-a4b6-31f1eedf762e.↩︎