Predictions for 2015

Making predictions or forecasts about a national or the world economy is fraught with failure. There are so many variables to consider and the data available are often inadequate and biased. But the main reason why economists are usually wrong with their forecasts and predictions is that mainstream economics has no real theory or laws of economics to build predictions on. Nevertheless, they carry on making them, either as an act of hubris (we are so clever!) or because their bosses and clients in investment banks demand it.

No wonder professional economists have predicted none of the last seven recessions. Mainstream economists, being an apologia for the success of market economies and capitalism, never predict a recession, even when one is staring them in the face, as it did at the beginning of 2008. Indeed, since 2001, only an average 12.7% of economists surveyed by James Montier from GMO were right about the following year’s economic growth.

Anyway, I shall try and make some predictions and forecasts for 2015. And at least my forecasts are based, I think, on relying on some important underlying laws of motion of capitalism. That should make my forecast a touch more accurate – I think. Economists, if they are serious about making a science of political economy (perhaps a contradiction in terms!), must make predictions as part of any test of laws, as in natural sciences. It is not good enough for Marxist economists just to say, well, ‘under capitalism there will be recurrent crises (slumps)’. We have to do better than that: namely, at what stage is capitalism going through; is it on an upswing or downswing; is a slump close to hand or some way away?

Back in 2005, when I wrote much of my book, The Great Recession (published in November 2009), I forecast that a major slump was likely to take place in 2009-10. This is what I said: “There has not been such a coincidence of cycles since 1991. And this time (unlike 1991), it will be accompanied by the downwave in profitability within the downwave in Kondratiev prices cycle. It is all at the bottom of the hill in 2009-2010! That suggests we can expect a very severe economic slump of a degree not seen since 1980-2 or more”.

As the quote suggests, I based this forecast or prediction on some laws of motion in capitalism that I had identified. The first was the long cycle in global production prices, namely a period of 27-36 years of general upswing in commodity and production trade prices, followed by a similar downswing. This is called the Kondratiev wave. I reckoned that since 1982, the Kondratiev cycle was in a downswing that could last until 2018. This would keep inflation low and indeed deflation of prices would appear, placing downward pressure on global investment.

The second was that the domestic construction or property cycle (of about 18 years) in major economies like the US and the UK was reaching its peak and we could expect a housing bust around 2009.

Third, and most important, I had discerned that there was profit cycle that could be identified for the major capitalist economies over about 32-36 years from trough to trough. From the late 1990s, most capitalist economies were experiencing a downwave in the profitability of capital that was only being propped up by a credit boom and fictitious capital profits. The downwave would come to a new trough about 2015-16. And in such a downwave, more frequent and deep recessions were likely, as they had been in the previous downwave of 1965-82: with a weak one in 1969-70; a major international one in 1974-5 and finally the double-dip slump of 1980-2. This created the conditions for a revival in profitability (the so-called neoliberal period) that lasted until the end of the century. Back in 2005-6, I reckoned the profit downwave signalled at least another huge slump.

world rate of profit

Finally, there was the Juglar cycle of growth, investment and employment which seemed to last about 8-9 years from recession to recession in modern economies. These recessions took place in profitability upswings, but then they were weak and scarce (1958 or 1990). In downswings, they were more frequent and severe (1929-32, 1937-8 or 1974-5, 1980-2).

On that basis, I reckoned that all these cycles would come together in a major slump around 2009-10, as we had not had all four cycles in a downswing together before since the 1930. Well, I was slightly wrong: the credit bubble burst in 2007 and the Great Recession came one year earlier than I predicted, in 2008-9.

As we enter 2015, both the Kondratiev and profitability cycles are still in a downwave, in my opinion, but the construction cycle has turned up (in some cases leading the current weak ‘recovery’). But another slump must still be on the agenda to enable sufficient destruction of capital values to deliver a new upwave in profitability and also see the end of Kondratiev downwave. If you like, the major economies are halfway through a Long Depression, as in the 1880-1890s or the 1930s. The 1974 slump was eventually followed with the 1980-2 slump; the 1929-32 slump was followed by a recovery and then another slump in 1937-8. So on that basis, after the recovery of 2009-14, we can expect another slump by around 2016-17. There – I have said it now.

But what of 2015? Well let’s start by reminding myself of what I said this time last year about 2014 (https://thenextrecession.wordpress.com/2013/12/30/faster-growth-in-2014/). My general forecast for 2014 was that, contrary to the optimism of the professional mainstream economists working for big money at the likes of Goldman Sachs, the ‘global crawl’ would continue i.e. global economic growth would continue to be weak and below average as it had been ever since the end of the Great Recession in 2009. And so it has proved.

low gear

Back at the beginning of 2014, the IMF forecast a 3.6% expansion in real GDP in the world economy. It will come in at just over 3%, well below the average. The rich investment bankers at Deutsche Bank and Goldman Sachs reckoned that the US economy would grow “above-trend” in 2014. Well, it’s true that the US had a good third quarter of growth, being the best performing capitalist economy since the end of the Great Recession (see my post, https://thenextrecession.wordpress.com/2014/12/24/us-economy-ends-on-a-high-the-uk-le), but it will still achieve only about 2.5% growth, at best, in 2014, well below average trend growth of 3.3% since the 1980s.

As for the UK, I reported last year that the Conservative-led government’s finance minister, George Osborne, was being lauded as a hero because the UK economy would jump ahead in 2014. And indeed, up until the latest data on growth, it seemed that the UK would achieve the fastest expansion of the top G7 capitalist economies in 2014, at 3%. However, such hopes have now been dashed and the UK economy will only manage about 2.5-2.7% this year.

Even this growth has been based on government stimulus for house-building and cheap credit. Investment in productive sectors has been weak. Real GDP per person is still below the peak achieved in 2008 before the Great Recession. Investment in real terms is still 4% below, manufacturing output 5% below and productivity per hour still over 1% below its peak before the Great Recession. Britain is running a payments deficit with the rest of the world equivalent to 6% of GDP, bigger than the government deficit. Above all, net national disposable income per head (after inflation) is some 6% below the peak and real weekly earnings have fallen since the Tories took office in May 2010 in every quarter but two – the longest fall in real wages since the Great Depression.

And it does not look any better in 2015. The housing boom in the UK is beginning to fade and will expose the underlying weakness in the productive sectors in 2015. Average house prices are rising at their slowest rate for more than a year. And according to the Chartered Institute of Personnel and Development (CIPD), UK wage growth is likely to remain weak for at least another year because employers are finding it easy to recruit and retain staff at current pay levels. Employers surveyed in the autumn had an average of 50 applicants for every low-skilled vacancy and 20 for every high-skilled vacancy, 40 per cent of whom were suitable for the role. They also reported that job turnover remained low. “If few employees are leaving, and most employers can find suitable applicants for vacancies, the conditions required for higher across-the-board pay rises are not being met,” said Mark Beatson, CIPD’s chief economist.

Indeed, that is the story for most capitalist economies in 2014 (Europe, Japan, the US and the UK): weak economic growth, poor business investment and falling real incomes for the average household.

Last year I forecast a very weak Japanese economy and it has proved to be even worse that I forecast, despite dollops of virtually interest-free credit, fiscal stimulus packages and another election to install Abenomics – the supposed answer to Japan’s woes (see my post, https://thenextrecession.wordpress.com/2014/10/13/japan-the-failure-of-abenomics/).

And lo and behold, as I write this, the Bank of Japan’s own economic activity survey, the Tankan, suggests that “Japan is going into a double dip”, recession with only a small quarterly increase in GDP for the final quarter of 2014 followed by a renewed contraction next year. In the survey, large Japanese companies gave a score of 14 for current business conditions and forecast a decline to 12 in three months’ time. For small companies, the respective figures were 0 and -4.

The Abe government has launched yet another round of fiscal stimulus measures. It’s going to cut corporate tax rates to try and encourage Japanese companies to invest more and raise wages for their workers and so boost demand. Some hope! Large companies have instead been building up their cash reserves and keeping wages down.

And all this extra spending threatens the other major policy of Abenomics – to get Japan’s huge public sector debt and deficits down. The debt has not been a problem up to now because the interest rate on that debt is near zero and Japan’s banks are pressured to buy government bonds. But this means that little productive investment takes place and if interest rates were to rise, Japan could face a serious debt crisis. The government wanted to reduce its annual deficit by half in 2015 and ‘balance the books’ by the end of the decade. But the economy has been so weak that it has been forced to hold back an increase in sales tax, spend $28bn on projects and give new handouts to corporations. The fiscal target is a joke. What is down the road will be a severe cut in welfare and social benefits to pay for this.

As for Europe, there is no relief. Greece may have finally bottomed the deepest slump in its modern history in 2014, but only after a 40% fall in average living standards and the destruction of jobs, welfare, health and public services. Now it is about to enter a major battle with the EU leaders over how to recover (see below). Spain is still recording record high unemployment levels and with little sign that the productive sectors of the economy are turning up; Italy remains in a depressed state. Only the German economy looks relatively better.

The global economy remains in a crawl and will do so in 2015 for one good reason: the failure of business investment to leap forward. Goldman Sachs reckoned this time last year that there would be a global investment boom in 2014. That has proved to be a mirage in Europe, Japan, the UK and in the major so-called emerging economies of China, India, Brazil and Russia, where investment growth has slowed markedly or collapsed, as in the case of Russia (see my post, https://thenextrecession.wordpress.com/2014/12/08/oil-the-rouble-and-the-spectre-of-deflation/). The emerging markets of Brazil, Russia, India and China collectively known as the BRICs — will likely grow in 2015 at their slowest pace in six years, according to Oxford Economics. Only the US has shown some pick-up in investment.

As I said last year, the reason that business investment has not boomed is that in most economies average profitability remains below levels before the Great Recession and below levels reached in the late 1990s. Most economies are still experiencing the downwave in the profitability cycle, as explained above. Coupled with the downwave in the Kondratiev cycle, that is why the global capitalist economy is in what I call a Long Depression, with some years to run.

Let’s finish with a few predictions.

First, the failure of the Samaras government to get its candidate for President elected by the Greek parliament has forced an election at the end of January that the leftist Syriza party is likely to win, assuming the public opinion polls remain as they are. Syriza will likely have to form a coalition with smaller left and centrist parties. It is pledged to renegotiate the debt burden that the previous government has built up with the EU, some €322bn. This can never be paid off and remains a millstone around the neck of the Greek economy and its people for decades.

greek debt

Syriza wants much of it written off. The EU leaders will play hardball, if only because it is making Ireland and Portugal pay their loans back in spades and it won’t want to set a precedent for other Euro debtors. The existing Troika programme funds are supposed to fund the Greek government from February as long as the government agrees to more fiscal austerity. Syriza claims it will do no such thing, so there appears to an impasse with Greece running out of cash by the summer.

I reckon that this process will spin out through the next few months with nothing resolved. There is a way out for the EU leaders if they want to keep Greece in the euro: they could eventually agree to a perpetual rollover of the debt – so the debt stays on Greek books but there’s nothing to pay for the foreseeable future. This could be sold as the ‘Greek exception’. Alternatively, Syriza reaches a compromise agreement to cover the debt. Either way, my prediction is that Greece will still be in the Eurozone this time next year.

What’s going to be interesting in 2015 is the number of elections in key peripheral Eurozone states coming up. Apart from Greece, Ireland and probably Italy will have elections, where weak centrist governments will try to ‘hold the line’ against populist parties. I will consider the economic programmes of leftist parties like Syriza and Podemos in future posts.

And then there is the UK with a general election in May. I have made three predictions about the UK in the past. The first was that Scotland would vote to stay in the United Kingdom in the referendum last September (see my post, https://thenextrecession.wordpress.com/2014/09/19/scotland-one-prediction-right). That happened. The second was that the Conservative-led coalition would be returned to office (probably the Liberals in tow again). And the third was that the British people would vote to stay in the EU if there is referendum on that in 2017. I’ll explain my prediction for the May election in a post nearer the date.

For 2015, I have launched a Facebook site so that you can keep up to date with my posts and other events and papers.  See
https://www.facebook.com/pages/Michael-Roberts-blog/925340197491022

and

Essays on inequality
Createspace https://www.createspace.com/5078983

or Kindle version for US:
http://www.amazon.com/dp/B00RES373S
and UK
http://www.amazon.co.uk/s/?field-keywords=Essays%20on%20inequality%20%28Essays%20on%20modern%20economies%20Book%201%29&node=341677031.

38 thoughts on “Predictions for 2015

      1. You said one of these days that to recover profitability a world war would be necessary. Or maybe I am confused?

      2. Not every depression has been ended by war. But there have not been many depressions to judge! Perhaps the first capitalist depression of the early 1800s was ended by the Napoleonic wars. And the Great Depression of the 1930s was ended when America entered the war in 1941 and built up arms before hand. The Long Depression beginning in 1873 did not end in a world or even European war so it actually lasted longer and with varying degree of severity in different countries up to 1896. The current Long Depression is only seven years old. I think it might end around 2018 if there is a sufficient destruction of capital values by then. All this is a calculated guess – but calculated. I have said the opposite to what you think I said. I dont think this current depression will be ended by a world war, although there is permanent war – regional and local ones.

      3. Michael, look at your graph, the long depression of the XIX century is finished by the 2nd world war.
        Observe that if you trace an average line up from 1868 to 1905, it intercepts the deep valley of 1938 very well.

        Since you are looking at the world rate of profit, It looks like the pattern of profit of WW1 begins around effectively 1908, that is after the Russo-Japanese war, if there is a war trigger for the profit, probably triggered by an intense world wide arms race.
        So, I’d say that in terms of profit recovery, Russo-Japenese war is the start of a unique World War that lasted from 1905 until 1945.

        How else you’d expect the enormous plateau above the declining line I described above, starting with a gigantic spike?

      4. Daniel – It’s an interesting idea. But I think it is better to eyeball the troughs and the peaks. If you do that then the peaks are at 1870, the early 1920s and the mid-1960s and 1997. The troughs (even more crucial) are end 1890s; mid-1930s; 1980-2; and still to come?. Characterising the Great Depression of the 1930s as part of some long profit revival from 1908 to the mid-1970s does not work for me. But it is worth a discussion in the future. I’m returning to this subject in early 2015.

      5. Michael, this is an eyeball approximation, but notice that there are very steep slopes towards the old trend of the XIX century. Se the slope crisis of the 1930’s and the one of the post war and in 1983. I think we need one that will throw the rate from the actual 20% to 10%+3%-3% (uncertainty due eyballing) in the depth of the valley

      6. Michael Roberts, but isn’t that a long wave? And that is the period of intense wars, we could say that the whole period was of a continuous world war. It’s just that we forget that Africa and South East Asia are not much worth considering as human being (shamely, I consider myself in that). But that was a boost in profit realization of military industry.
        Nowadays, the military sales do not circulate as much, since they are not as much destoryed, and its profit sucks from other sectors. See that the most advanced planes from US are plagued with defects and that they are close in power to 30 year old Russian airplane projects.

      7. Michael, the data on Esteban’s paper is not the same as the graph you posted above. There is no graph there with the spike around 1905-1910, from 30-45%, neither in the data presented at the end of the paper.

  1. PS: We’re an independent Marxist blog based in New Zealand. If you’re ever down this way, do get in touch. . .

    Phil

  2. Nice article, Michael. Late 2015 will also see a general election in Spain, the runup to which looking like being dominated by the rise of Podemos. With regard to the latter, what happens in Greece will be significant.

  3. I subscribe to Ernest Mandel’s interpretation that you need exogamous factors to launch a new expansionary long wave. Not necessarily a world war, although none of us should be ruling that out.

  4. Just something I think you need to take account: the fracking oil at USA will burst this year, they need oil be at US$ 100 for be profitable, oil bellow US$ 60 and going to be US$ 40 they will go broke. That will bring a huge unemployment wave and a big recession, and it is possible that it will bring too a new financial crisis.

  5. Here’s a prediction: growth will slow, turn negative, etc; things will get worse, the rate of profit will fall, and capitalism will not collapse; it will go on being capitalism until it is overthrown. Overproduction, financial crises, plummeting oil prices are just part of capital being capitalism; “The price of beef may be high or low, but it always involves the same sacrifice for the ox.”

      1. Well, the eurozone is of course the big unknown and I think the ECB will be forced to act early in the new year. If there is any improvement in the eurozone, global growth will probably be higher next year.

      2. OK, that’s your prediction. I have two predictions: (1) This shit–capitalism– will continue until it’s overthrown. (2) Predictions will make absolutely no difference.

  6. Ok, you predicted the slump. But did you predict that the USA government would bail out the banks, or GM? Where are these structural realities factored into the analysis? If you stick rigidly to Marxist theory you can miss the important details, which are not surface appearance but fundamental.

    Other than that I can’t disagree with much you say here.

    1. You don’t need to be a Marxist to have predicted that– that the govt would bailout the banks and/or GM. Govt had bailed out Chrysler, Citibank, the S&L industry in prior episodes, so you don’t really have to have access to the MEGA to figure something was going to be done. That’s what the government is there for– to bail out the CLASS.

      Now could anyone have predicted that the Fed and the US Treasury would have let Lehman Bros. sink and save AIG?

      I don’t know, but it certainly isn’t a problem with Marxist analysis if it doesn’t predict every single move the bourgeoisie make or don’t make.

      1. Fair enough, you certainly can’t predict every last policy decision.

        But the point is that these taken cumulatively have a
        huge structural affect, which tend to be ignored when Marxian theory is applied to the analysis (This is more a view of Marxian theory rather than marx himself). And these things that you claim are easy to predict must mean they are more than random policies but something fundamental to capitalism. Well if that is the case Marxian theory doesn’t really account for them, certainly not viewed through the closed and abstract model that constitutes the 3 volumes of capital and crisis theory in general.

      2. Now that’s very interesting. I think maybe the problem is in viewing Capital as a “close and abstract” model– rather than as a critical analysis of the laws of capital accumulation and the immanent contradictions of that accumulation.

        In a sense– right, no abstract closed theory can account for all variations, iterations of structural changes in a concrete mode of production. But historical materialism can and does.

        I think that’s what we’re “supposed” to do with Capital– grasp its analysis as a means for apprehending what is going on here (or there) now (and then) in the mode of production.

        Capital is not complete, in, of and by itself, without that historical materialism– without that recognition and apprehension of class struggle.

      3. Yep, I agree with that. Except closed and abstract can be at the same time a critical analysis. i think Marx was fully justified in his abstractions in Capital. I just think some Marxists can’t see the world and the system beyond the abstractions.

    2. Some Marxists use a high frequency crisis predictor so it’s not surprising that they sometimes get it right. Predicting capitalist non-crisis periods is less successful. And slow, stagnation, crisis and depression get used too loosely.

      1. Maybe, but 2008 was a crisis! Some think it was merely a blip.

        There is one thing not predicting a crisis but if you can’t actually see one when it explodes in front of you then you have to ask, what use are you? I think the permanent revolution perspective, which I assume you are part of (Billj and that crowd), didn’t regard 2008 as a crisis at all?

      2. Graham,

        I’m not sure what you mean: Do you mean that the Fed/Tsy decision to rescue AIG and let Lehman Bros sink was the determining factor in the breadth and depth of the crisis?

        If so then we’re back to the “mistaken policy” explanation for the the movement of capital; and of course this is the argument of political economists of all stripes from Friedman to Krugman and all those in between.

        And of course, the corollary to the mistaken policy argument is the “crisis of confidence” argument, as in “all we need is confidence; the only thing we have to fear is fear…” ad infinitum ad nauseum.

        What made “it”– the crisis of 2008 “big”– was what made and makes capitalism big– accumulation, overaccumulation, the tendency of the rate of profit to decline, the overproduction that is part and parcel of the overaccumulation.

        Lehman Bros failure made it big? Not hardly. Lehman Bros failure was the result of how deep, and broad, the crisis was.

        How about Northern Rock? Did that “make” it big? Dexia? The various landesbanks? Countrywide? Washington Mutual? FNMA? FMAC? Ango-Irish? Greece? How about Greece? You think that might have been a bit bigger than Lehman Bros? RBS? Iceland?

  7. How did we get here and where will we go?

    Behind every crisis of capitalism is the fact that it takes more and more capital to make a decent rate of return. Capital constantly comes up against its political economic limits which prevent further growth and even throws the system into reverse. It has to burst through these arrangement once so favourable if it is to grow again. It takes great violence to overthrow the previously established political economy. It took two world wars and a Great Depression to establish American hegemony over a declining Europe and even then the assistance of world Stalinism was required. But the post-War political economic arrangements could only deliver a boom of less than thirty years duration before profitability became an issue once again. This crisis of profitability which left the West stagnant, sclerotic and monopolised forced it into seeking an aggressive victory in the Cold War. An assault on the domestic and international working classes was launched along with an unprecedented credit-driven boom based on bank de-regulation. Stalinism surrendered in the teeth of the West’s spending and its weaponry not to mention a Western leadership seemingly mad enough to use them. Victory pushed America to seek to establish itself as the sole and only global super power. But the mighty US became militarily and morally bogged down in Iraq and Afghanistan and in 2008 the real cost of `winning’ the Cold War became apparent when the Bankers’ 30-year Ponzi Scheme collapsed in spectacular fashion. The very forces whose emergence US imperialism had hoped to prevent were now evoked by its own efforts and US-sponsored globalisation began to unravel. The world economy is now stagnant, sclerotic, monopolised and irrevocably and completely bankrupt. The unravelling of globalisation will be a process ten times more violent than the process that established it and of course there is no America waiting in the wings to save America. There are no seriously conceivable new political economic arrangement into which capitalism can now move even if the old is swept away with great violence. A system that cannot change is a dead system. Capitalism is a dead system. Our choice is to go with it to a New Dark Ages in which war is a permanent condition, a regime of global savagery a glimpse of which can be seen in Syria or Gaza or Baghdad today, or we can transcend capitalist globalisation through world proletarian revolution establishing a global commonwealth of proletarian nations taking global economic integration to the next level and putting our barbaric past behind us.

  8. “So, I’d say that in terms of profit recovery, Russo-Japanese war is the start of a unique World War that lasted from 1905 until 1945.”

    Yep,that’s a good starting point. Russia (Federation) and Japan, two sinking countries today.

    Thanks for your mighty efforts generally, Michael.

    Don’t know about Kontratief waves myself, but I agree that the end of the long depression involved an escalating series of imperialist wars, culminating in the First World War. That’s not likely this time around, despite that destruction of the Russian Federation is virtually official U.S. policy. The Russian neo-con far right and fascist fringe cannot be mobilized for the destruction of the multinational federated state, as they presently have in Russian neo-imperialism what they would seek for if they were Ukrainian. The Russian neo-liberals cannot overthrow the Putin regime, or any possible Russian Federation regime, without the Russian neo-fascists, as the actual outcome of the Maidan showed. And Ukraine is scheduled for reaming this year, by the EU and US oligarchs. Anyway a couple of key predictions neglected.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.