Martin Wolf and what to do about the commodity shock

Martin Wolf's column today reflects on the commodity shock on the basis of the data provided by the World Bank in the last Commodity Markets Outlook. As has been the prevailing consensus he suggests that this is an unprecedented shock, larger than the oil shocks of the 1970s, and any other commodity crises of the last 50 years.

In his words, based on the World Bank: "the initial impact of the closure of the strait was a global loss of 10.1mn barrels a day of oil in March. This was much larger than the impact of the Iranian revolution in 1979, the Arab oil embargo in 1973, Saddam Hussein’s invasion of Kuwait in 1990, or the Iran-Iraq war in the 1980s."

The current data on the price of oil has not yet confirmed these pessimistic predictions. The issue might be related to how much of a chockepoint the Strait of Hormuz really is, and how trade can adapt to it. Note that between 1967 and 1975 the Suez Canal was closed and forced oil shipments around the Cape of Good Hope. This caused a surge in transport costs, incentivized the development of massive supertankers, and forced a rerouting of trade, but it had limited initial impact on oil prices compared to the 1973 oil crisis.

It is obviously to soon to say, but so far the current oil shock has not lived to previous ones, when oil producing countries in OPEC had more power. Figure below shows the nominal and real price of Brent oil, which remain quite below the 1970s peaks.

Wolf is on the pessimistic side of things, and suggest, correctly on that, that costs will be unevenly distributed and that action to avoid worst problems is necessary.  He says: "this is a sizeable disruption, which is sure to hit many of the world’s poorest people and most vulnerable countries hard. The rise in oil and fertiliser prices guarantees that. This underlines the moral case for continuing to provide international assistance." That's all good, even if one is more cautious about possible outcomes.

But then he goes on to say that: "central banks are going to have a tricky time navigating the consequences. But they must not let inflationary expectations slip out of control." In other words, this is going to hurt the poor, and he thinks it will very much, since this is an unprecedented shock, but central banks should not care, and be concerned only with inflation expectations.

Before the Pandemic inflation there was some growing understanding, now less so, that inflationary expectations are not that relevant (on that see this). Jeremy Rudd wrote an article for ROKE suggesting that modern macroeconomics assigns too much importance to inflation expectations, with weak theoretical and empirical justification. Measures of inflation expectations (surveys, market-based) are noisy, inconsistent and weakly correlated with actual inflation outcomes. For him, this makes them unreliable as a key driver in models.

If inflation is caused by a hike in costs of production, there is very little that a hike in interest rate could do to control it, other than repress wage increases by slowing down the economy. Not that this should be a major concern in most advanced economies, at least.

My guess is that the impact will be smaller than expected, and that central banks will overreact. That might be the case of the Fed today, that should reduce rates, but probably won't because of inflationary expectations.

Latin American Corner: Quantitative easing, commodities, corporate debt and the paradox of debt

By Naked Keynes (Guest Blogger)

The policy of quantitative easing (QE) pursued by the Federal Reserve following the fall of Lehman Brothers in September 2008 meant to lower long-term interest rates in the United States and boost expenditure had major effects on developing economies including in those of Latin America. As it is well know QE did not increase liquidity. The liquidity with which the Federal Reserve bought financial assets ended as excess reserves at the Federal Reserve balance sheet and the money multiplier became, actually a divisor (the money multiplier dropped below 1 after the start of QE).

However, quantitative easing had an important portfolio rebalancing effect, which altered the relative profitabilities of different assets and made commodities an attractive investment and speculative alternative. Investing in commodities as a way to hedge risk was also championed by mainstream economists. Gorton and Rouwenhorst (2004) (here) argued that commodities and stocks yield similar returns over time so that they are adequate investment substitutes. Moreover they claimed that commodities and stocks are, in terms of levels and volatilities, either not correlated or negatively correlated over time so that investing a part of the portfolio in commodities lowers its total risk. The effect of the boom in commodities during the 2000s in Latin America is reflected in the difference between GDP and gross domestic income (GDI) (the latter is equal to GDP plus the effects of the term-of-trade which were clearly favorable in many Latin American economies. See figure 1 below for a comparison of GDP and GDI for major Latin American commodity exporters.

This gave a false sense of prosperity that even endured after GDP growth in most economies of the region began to decline in 2011: domestic income follows GDP but with a lag).

Moreover, the increase in commodities and terms-of-trade was accompanied by nominal and real exchange rate appreciation in the great majority of Latin American countries. The appreciation in the exchange rate was detrimental to export diversification but, nonetheless, welcome by policy makers as it permitted to control inflation.

Low external interest rates, high commodity prices and income, and exchange rate appreciation set the stage for the large rise observed in the corporate debt following the Global Financial Crisis (2008-2009) in the larger economies of the region. The issuance of corporate debt involved some of the major firms in the region and major commodity producing firms as well the financial sector.

According to recent estimates (here) corporate bond issuance in foreign currency increased from US$ 170 to 383 billion dollars between 2010 and 2015 (compare with FDI figures which are of the order of US$ 150-160 billion per year). But these corporate debt estimates are flows. The corporate debt stocks are much higher. Available non-official figures for Brazil point to a corporate debt stock of US$ 300 billion dollars. The corporate debt issue not only affects Latin America but in general the developing world. According to a research note published by the Institute of International Finance (IIF, June 1 2015), the stock of corporate debt outstanding was above US$ 6.8 trillion in 2014.

The decline in commodity prices which began in 2011 and turned into an outright crash in 2015 was accompanied by exchange rate depreciation and lower growth. The decline in commodity prices reduced the income flows of tradable sector firms giving rise to a wedge between income flows and debt flows. In the case of the non-tradable sectors, the decline commodity prices plus exchange rate depreciation had both income flows and balance sheet effects. Firms have reacted by reducing the debt issuance and slash capital expenditure which affects negatively investment and growth.

As argued by Fitch “the refinancing risk is high and default risk in climbing” (Fitch Latin American Corporate Bond maturities, February 8, 2016). Moreover, as firms have tried to cut investment, this has affected profits so that the corporate sector has increased its debt instead of deleveraging. This paradox of debt, prevalent in part of the Post Keynesian literature (See, M. Lavoie, New Foundations of Post Keynesian Economics, 2014) can be illustrated in figure 2 above with the case of Chile and may be a major stumbling block to overcome the current slowdown of some of the major Latin American economies.

Global depreciation since the collapse of oil prices

One figure is worth a thousand words (negative number imply depreciation).

So, if you think China devalued a lot...

Source here.

PS: On Twitter some have argued that some depreciations are overestimated. My point remains. China did not depreciate much, comparatively. And besides as noted before there might be a connection between depreciation in the periphery and lower commodity prices.

Financialization and the Resource Curse in Brazil

"Financialization and the Resource Curse: The Challenge of Exchange Rate Management in Brazil"

By Kevin P. Gallagher and Daniela Magalhães Prates

Indeed, Brazil has been blessed and cursed with high commodity prices (from 2003 to mid-2008 and 2009-2011) and low interest rates in the core economies after the 2008 global financial crisis. Such an environment, coupled with the high domestic policy rate and the sophistication of the Brazilian financial system, has made Brazil a much sought after destination for carry trade operations through short-term financial flows that are largely transmitted through the foreign exchange derivatives market. Speculative operations into this market have accentuated the upward pressure on the exchange rate, which has come with higher commodities prices, leading to what we refer to here as a financialization of the resource curse (pp. 2).

Read rest here.

CEPR: Latin American Growth in the 21st Century - The 'Commodities Boom' That Wasn't

By David Rosnick and Mark Weisbrot

This paper looks at whether the data support such a conclusion. It finds that there is no statistically significant relationship between the increase in the terms of trade (TOT) for Latin American countries and their GDP growth. There is, however, a positive relationship between the TOT increase and an improvement in the current account balance. It may be that this allowed countries to avoid balance of payments crises or constraints.

Read rest here.

Commodity dependence in comparative perspective

Last weekend at the Eastern Economic Association Meetings in Boston Bilge Erten presented her work, together with José Antonio Ocampo, on the commodities super-cycle (see here). The figure below (click to enlarge) was on her power point (not in the paper) and is from UNCTAD's State of Commodity's Dependence 2012.

Note that much of Sub-Saharan Africa has levels of commodity exports as a share of total exports of between 80 and 100%. South America is not too far. China has only 8%, even if one can arguably suggest that some of the manufactured goods exported by China, which are highly standardized, are commodified. That would also explain why Mexico and some Central American economies depend less on commodities, i.e. the export of manufactured goods from 'maquilas.' South Korea is also low at 12%. A few Asian countries do better than most of Africa and Latin America in this respect. That is, they would be less vulnerable to changes in commodity prices.

The super-cycle of commodities is NOT over, it's just not super anymore

Or at least is what The Economist says (here). Graph below shows a very long series for commodity prices.

Note that the cycle (super or not) is less visible than the trend (hints of Prebisch-Singer). At any rate, the fall in 2009 and the recovery and then slow fall after 2011 imply that the overall index is now higher than at the lowest point in the series, the late 1990s, and around the same place it was in the early 1950s.

Is Argentina on the verge of an external crisis?

There is for starters the question of what causes external crises. As I have noted in other places (chapter 7 here or here, for example), external crises are NOT caused, in general, by fiscal deficits (quite the opposite, fiscal crises are the result of balance of payments crises). External crises result from the inability to service foreign debt (and to import intermediate and foreign goods), which are caused by a shortage of foreign currency (i.e. dollars).

As it can be seen in the graph above (data from Orlando Ferreres for those concerned with the sources), the current account surplus as a share of exports has shrunk and is now negative (at around 4% or so of exports). Note, however, that the level is far from desperate, and well below the crises levels when the current account deficit is above 60% of the exports.

Part of the anxiety is associated to the fall in the central bank's reserves, which stand at around US$33 billions now, down from slightly more than US$50 in 2011. The European crisis and the negative real rates of interest explain the drain on reserves, which are also not at a critical point right now. A combination of exchange controls, that have been in place (and have not been particularly efficient), and higher rates of interest might stop the outflows.*

Sure enough a balance of payments crisis could ensue, if say Vulture Funds eventually force a default, or if an external shock like a worsening of the crisis in the central countries followed by flight to safety, or a collapse of the terms-of-trade lead to a sudden decrease in the value of exports. But those do not seem to be necessarily intrinsic to the Argentine situation, and a slow recovery in the center, with significant amounts of international liquidity, and no incredible collapse of the prices of commodities seems as likely as the alternative.

In other words, the problem in Argentina, which is relevant for many countries in the region, is the long-term development strategy, and not the short-run balance of payments position. What the shrinking of the current account surpluses, and the resulting constraints on policy space, suggests is that the continuous dependence on commodity exports (manufacturing exports go mostly to the region, i.e. Brazil, and produce a deficit), and the absence of a more coherent policy of import substitution and of industrial development, continues to be relevant, as predicted more than 60 years ago by Prebisch and ECLAC.

* Higher rates can be compensated by subsidized credit by the public banks if demand for credit increases, but that would require demand expansion.

Commodity Prices

The Economist has posted a chart with some evidence of falling commodity prices, noting that since last September their index has fallen by 12%. The most discussed fall in recent times had been that of Gold, but others are relevant too.

Note that this masks significant variation in commodity price dynamics. For example, soybean prices fell from US$ 615 in August to 536 last March. Still high by historical standards, but considerably lower. The same is true for copper, with a fall from US$ 8,087 last September to 7,652 in March. Natural gas, on the other hand, went from around US$ 70 in March 2012 to about 136 last March, which is not the highest from a long term perspective. This suggests that commodity producers will be hit very unevenly by variations in prices, as will importers.

PS: All data available here.

It's NOT just the external conditions

Frankly, it's a bit boring. But every time you show the actual data on Argentina's growth someone says it's just good luck. The terms of trade boom. Data below compares terms of trade in Argentina and Brazil in the same period, 2003-12 (Kirchner-Kirchner and Lula-Dilma).

Not very different, and if something terms of trade improved a little bit more in Brazil (33% against 22% approximately, for the whole period). Note that metallic commodities (Brazil is a big exporter of iron ore) increased more than agricultural goods (both Brazil and Argentina are big exporters of soybean and derived products, for example). So did Brazil grew more than Argentina, which is what you would expect if external conditions determined growth? See graph below.

As it turns out Argentina grew 5.8% against 3.6% in Brazil. Further Brazil has a larger current account deficit. Fiscal and monetary policy were more expansionary in Argentina, and the nominal and real exchange rates more devalued. Let alone that the average real wage grew more in Argentina. So no it was NOT all external conditions.

Can we please, pretty please, with sugar on top, stop just venting prejudices and get the data before we talk. You are entitled to be against the policies of the government, you might even be for the policies of the Washington Consensus (people have defended crazier ideas). But, as Daniel Patrick Moynihan used to say, you are not entitled to your own facts.

Industrialization, Wages and the Terms of Trade

The quote above is from Raúl Prebisch's classic paper "The Economic Development of Latin America and Its Principal Problems," the so-called Development Manifesto (available in Spanish here), published in Spanish and Portuguese in 1949, and the following year in English.

Note that his explanation for the tendency of terms of trade of commodities to fall over time (the so-called Prebisch-Singer hypothesis) was based on the fact that in the boom wages went up in the center, but not so much in the periphery, since industrial workers in the center were organized and could demand higher salaries, while that was not possible for the agricultural and mining workers in the periphery. So in the recession, while prices of commodities and wages fell in the periphery, they didn't in the center. Class conflict, and not just technological change, was at the heart of the asymmetries between the center and the periphery.

Hence, industrialization in the periphery, and the re-organization of the labor force, would imply that more workers in the periphery would be able to keep part of the benefits of higher productivity. Industrialization would be good for the production of the commodity sector, since prices of commodities would go up, with higher wages in the periphery.

Also, note that this explanation of terms of trade suggests that if wages in the periphery fall, then the prices of commodities fall too. And it is worth remembering that a depreciation of the currencies of peripheral countries implies lower wages.

State of the World Economy: A South Centre's Perspective

During last Rio+20 Conference the South Centre organized a session with some local economists on the state of the world economy, which to a great extent reflects the views of their chief economist Yilmaz Akyüz (hat tip to Butch Montes for the link). Their views tend to be very pessimistic about the possibilities of sustained growth in the periphery and the continuity of the so-called double speed recovery, fast in the periphery and slow in the center.

For them the boom in the periphery in the New Millennium was caused by the commodity boom, and that, in turn, was dependent on China's exceptional growth record, which was heavily dependent on exports to developed countries. So the castle of cards is about to fall.

My views are slightly different (see here and here). I tend to think that Chinese growth may very well slow down, but may continue on the basis of domestic demand, and the structural transformation of the economy that will have to incorporate hundred millions in the next decades in the urban centers. Also, although the commodity boom eased the external constraint in other parts of the periphery, a lot of the growth was the result of the expansion of domestic demand too.

Also, as noted by the last Trade and Development Report by UNCTAD the expansion in several parts of the periphery has been based on higher rates of wage expansion, in other words, it has been based on an improvement of income distribution and higher domestic demand, in contrast to the stagnating wages in the center.

Finally, it is important also to note that commodity prices may also respond to speculation (as noted in the TDR report above), which explains the volatility of prices in the last decade, but also, as noticed in a recent paper by Franklin Serrano the supply conditions, and may for that reason be less vulnerable to a slowdown in China. In his view, part of the explanation of higher prices is the result of the revival of natural resource nationalism in a large number of developing countries, which has increased the state control of oil reserves and substantially raised the royalty rates and, the absolute rent component of the price of production of commodities.

So, perhaps China will still grow at a reasonably fast rate, and commodity prices might continue at relatively high levels, and there will still be some space for developing countries that do not decide, by doing fiscal adjustments of their own (yes I'm thinking about Latin Americans in particular), to cool down their economies to continue to grow. Sure enough a collapse of the euro, and of world trade, and a run for quality (towards dollars) might as well, as Abraracourcix feared, imply that the sky will fall on our heads, par Toutatis!

Gravitation, Full Cost Pricing and Prices of Production

Franklin Serrano (Guest blogger)

Most Sraffians understand that gravitation of market prices to normal prices is much quicker than the slower, but inevitable, adaptation of capacity to demand. But other eminent Sraffians have made some confusion by wrongly identifying classical prices of production with full cost pricing.

Classical prices of production are the centre of gravitation for market prices and are determined by the costs of the dominant techniques (at the level of normal utilization of fixed capital) and the state of distribution. It is a general theory of the structural determinants and limits for the trend of market prices in all types of markets. In spite of the similar name it has little or nothing to do with “normal cost” or full cost pricing which is a generalization of the descriptions given by some firms as to how they actually calculate their own prices based on a markup over their own costs (not those of the dominant technique).

First of all, there is obvious fact that the theory of prices of production was developed in a historical period in which such these pricing rules simply did not exist (see Hicks’ Market Theory of Money, 1989). And prices of production can still explain, in my view, the structural or trend element even in markets with highly flexible prices subject to wild short run fluctuations and rampant speculation, as in the so-called “commodity” markets (in Garegnani’s comment on Asimakopulos he explicitly mentions the importance of explaining the trend of the relative price of copper even though “at any one time copper prices are 50% or more above or below trend”)

Second, even in the so called fix price markets, were firms set the prices of their products directly, the full or "normal cost" that particular firms use to calculate their own price is the actual cost of these particular firms and the markup these particular firms think they can add to prices without trouble. These calculations generate actual market prices or (if stylized enough to have some generality short run theoretical prices) that are not unique even for a single market as the full cost prices can be different for different firms. These prices differ from prices of production because they refer to the actual costs of some firms and not the costs of the dominant technique available. For that particular product that determines a single price of production for that market.

The way prices of production may regulate the full cost prices of firms is by getting them in trouble whenever their actual costs plus their desired markups are too high relative to the costs (including normal profits) of the dominant technique, thereby attracting new entrants or cause some rival firms inside that market not to follow price rises that are due to increase in costs particular to that firm or “excessive” desired markups of these firms.

Professors Fred Lee and Marc Lavoie are both absolutely right and some Sraffians wrong in saying that full cost pricing is NOT the same thing as the classical theory of prices of production. Where I think they are definitely wrong is in thinking that classical prices of production are thus irrelevant for market forms in which firms follow such rules. For, through the power of actual or potential competition, the classical prices of production are the centers of gravitation that regulate even the trend of the prices of firms that practice full cost pricing. The closest analogy between classical prices of production and the industrial organization literature is thus the concept that Sylos-Labini called “limit” prices.

So market prices in both fix and flex price markets gravitate, towards or around classical prices of production. Any theory of full cost pricing can at best be a particular theory of short run price behavior of some firms in particular types of markets. There are old papers by James Clifton that started this confusion many years ago in Contributions to Political Economy and the Cambridge Journal of Economics. It is about time we stop confusing ourselves and our post Keynesian friends on this issue.

References:

Lavoie, M. (2003), “Kaleckian Effective Demand and Sraffian Normal Prices: Towards a reconciliation,” Review of Political Economy, 15(1) available here.

Lee, F. and T-H. Jo (2011) “Social Surplus Approach and Heterodox Economics,” Journal of Economic Issues, 45(4) available here.

Garegnani, P. (1988), “Actual and Normal Magnitudes: A Comment on Asimakopulos,” Political Economy, republished in Essays on Piero Sraffa: Critical Perspectives on the Revival of Classical Theory, Routledge, 1990.

Commodity prices and inflation

Krugman shows that according to a paper on the Chicago Fed, there is no evidence of a connection between core inflation (without energy and food prices) and commodity prices. He should have said no evidence in the United States.  In fact, when you look at developing countries the connection might be stronger.  That depends on the pass-through effect from commodity (tradables) prices to non-tradables.

Also, note that the Consumer Price Index (CPI) would be affected.  For example, if we look (see graph above) at the Argentine data (using terms of trade and consumer prices) there is a clear effect of commodity prices on CPI inflation.  That does not mean that I would favor hiking rates of interest, and putting a break, in the US or Argentina for that matter, on expansionist policies.  Quite the opposite, the suggestion is that policies to stabilize the wild fluctuation of commodity prices, for starters try to control the speculation with commodity derivatives, might be a good idea.