The Power and the Money

Mexico is gaining jobs, not wages

In 2009, in the wake of rapid inflation, Chinese unit labor costs were 72% of Mexico’s. (See page 19 at the link.) By early 2010, Chinese unit labor costs had risen to 86% of the Mexican level, and Mexico started to export cars to China. For some goods, landed costs (i.e., the cost of manufacturing and moving to the mainland U.S.) are much cheaper in Mexico than in China.

If you only look at the Chinese side, there seems little prospect of these trends turning around. China is experiencing bad labor shortages. Moreover, China has an inflation problem (the strategies described in this post seem to be hitting their limit) and its low inflation of 3% is not much lower than Mexico’s higher-than-expected 3.9%. Meanwhile, the peso is falling, while the yuan is likely to head nowhere but up. (If that turns out to be wrong, expect bad political consequences.)

On the other hand, if you look at the Mexican side, the reason Mexico is gaining competitiveness is that wages are flat. And flat at a low level. The good news is that Mexico is seeing rapid employment growth (see chart below) but Mexico no longer has endless millions of peasants to pull into the cities. Job growth is a good thing, of course, but it can go only go on for so long. (Mexico rebased its employment measures in 2010; thus the two indices.)

Translation: this is good news for Mexico, but it is a one-time thing. Without productivity growth, either within industries or by shifting into new products, Mexico wll remain relatively poor.

Implication: if the U.S. economy ever picks up again, so will emigration.

Omission: the effect of violence and extortion. The jury is out.

El perro que no ladra

 One strange thing about the upcoming Mexican election is how little the economy seems to matter. I don’t mean questions of long-term growth or distribution; those are very much an issue. I mean short-term business cycle stuff. Manufacturing output is up. The peso is down, but the central bank seems relaxed. Judging from the results on inflation-linked bonds, so are the markets. Consumer confidence has followed suit. Nobody is running a campaign like the ones in the United States or Europe. Growth since the 2008 shock is slow, but steady, as the below chart of GDP (in millions of 2003 pesos) shows.

To be frank, I was worried three years ago. But the bad economy dog did not bark. Why not?

Well, I was worried that Mexico was going the austerity route. But it didn’t! Mexico was no paragon of countercyclical fiscal policy, but it avoided anything that could be called austerity. I wrote the above entry in the third quarter of 2009 ... during which government spending was essentially flat, much less than the programmed 0.7% drop. (See below chart, which shows growth in GDP expenditures between government purchases of goods and services and everything else.) In short, my fear didn’t materialize because the PAN government never followed through with its own austerity rhetoric.

Still, simply not being stupid isn’t really enough to explain Mexico’s quick recovery. Lots of countries failed to implement sufficiently large stimulus packages and suffered for it. To explain Mexico we need to turn to two other Mexican strengths and a bit of good luck.

The first strength is, paradoxically, also a weakness: Mexico had a terrible terrible financial system going into the Lesser Depression. But from the point of view of protection from the effects of a worldwide financial freezeup, it was a great thing to go in with banks that did not lend and firms that did not borrow. (I do not recommend this as a long-term growth strategy.)

That meant that the Lesser Depression hit Mexico as a good old-fashioned trade shock. The below graph measures quarterly percentage changes in the final demand for Mexican production. (It is not the same as GDP, since GDP nets out imports.) You can watch exports plummet in the last quarter of 2008. Net exports barely budged (in fact, they grew) but all those unemployed export workers dragged down private consumption and all those hurting export manufacturers stopped investing. Thus the vertiginous decline in GDP in the first quarter of 2009 that you can see in the above two charts. (Demand continued to fall in the second quarter of 2009, but manifested as a drop in imports, leaving GDP flat.)

But this brings us to Mexico’s second strength: a flexible exchange rate. The peso plunged from a little under a dime to a low of 7¢ before recovering. (In point of fact, the Banco de México intervened to stop the fall, too soon in my opinion.) By the third quarter of 2009, exports had started growing again. By the second quarter of 2010, they surpassed their previous peak. In point of fact, Mexico became even more export oriented than before: exports grew from 26% in the middle of 2008 of GDP to 35% by the middle of 2012.

And that brings us to the good luck. Any guesses before I post? It seems to have been in the news lately.

Guest post: My problem with AMLO

 This is Patrick Corcoran, from Ganchoblog.

In general, I don’t find the arguments that Andrés Manuel López Obrador (aka AMLO) is a grave economic risk to Mexico to be very convincing. His program isn’t particularly extreme. Rogelio Ramírez de la O (his proposed finance minster) is sharp, experienced, and within the economic mainstream. And most importantly, between its business community, central bank autonomy, and NAFTA, there’s something of a policy straightjacket in Mexico. I don’t think anyone should expect a Kirchner a la mexicana should AMLO pull off the upset, and if I were a fund manager or a currency speculator, his election alone wouldn’t do all that much to change my opinion of Mexico’s prospects.

The problem is this: AMLO’s democratic commitment has wavered time and again. This is something that he has demonstrated repeatedly over the past six years. The reaction to the 2006 loss was a big part of that, but it’s not the only example. The takeovers of Congress were shameful. The shenanigans in Ixtapalapa treated the democratic process in an area of almost 2 million residents as though it existed only to serve him. And there’s no reason to think that this would change. When asked if he would accept the results of the election in July, a question where the only acceptable answer is an unqualified “Yes,” he waffled. His moderation during the campaign has been laudable, but when faced with a narrow defeat of an agenda item he holds precious, I don’t believe that he’ll just take his lumps and move on. Unfortunately, taking your lumps and moving on is a basic element of democracy.

The counterargument to this is that AMLO has been justified in his more extreme actions — that is, the election was a fraud, and the oil reform proposals being tossed around were a vital threat to Mexico’s well-being. Clearly AMLO seems to believe that; he couches his reaction in 2006 as a defense of democracy, notwithstanding his sending the institutions to the diablo. Given the fraud in the 1994 Tabasco race and President Fox’s 2005 push to deprive him of his legal immunity when he was mayor of Mexico City, I can understand him assuming any opposition to him is illegitimate. But while this bias is understandable on a personal level, that doesn’t make it any less worrying or damaging. Over the past six years, there isn’t much to support his belief that he was acting in democracy’s defense: the case for outright fraud in 2006 is extremely weak and has been contradicted by numerous people on the left, and his opinions regarding Pemex are just that — opinions — and have no more inherent value than those who would privatize the company tomorrow (which is certainly not what I’m advocating). The “desperate times, et cetera” explanation for AMLO’s unorthodox actions over the past six years just doesn’t hold up, and if your bar for the barbarians being lined up at the gates is set so low, well, then, what crushing political loss doesn’t justify an assault on the system?

That’s not to imply that his opponents would necessarily make better leaders. There’s also a pretty good argument to be made that the (potential) insidious erosion of democracy from within under Peña Nieto would far more harmful than AMLO’s frontal assault, which has typically been conducted in plain view of the public. If Peña Nieto turns out to be as bad as many assume he will — that is, if he is an old-school PRI dinosaur in a pricey suit — the damage could well be much worse than what is at risk in an AMLO presidency. Even if that doesn’t happen, by all indications Peña Nieto is the lightest of lightweights, and the campaign doesn’t seem to have put much weight on his bones. And the defects of Vázquez Mota are so obvious and damning that they hardly bear mentioning. From my point of view, there is no right candidate, and this is the worst slate of candidates of any Mexican election that I am more than passingly familiar with (basically, from 1988 onward).

But now that AMLO has emerged as the optimistic, anti-Peña Nieto candidate, it’s important to be clear about what his drawbacks are. Support for AMLO carries risks that go well beyond ideology. Electing AMLO would mean entrusting the system to someone who doesn’t wholly believe in it.

Argentina beats ICSID! Seriously, Argentina beats ICSID. Regularly.

In April, I had the honor to attend a conference on international investor-state arbitration held at the London School of Economics. (Below is a picture of me on the roof of the LSE after the conference. Yes, the weather was as bad as it looks.)

There I had the good luck to meet a law professor from William and Mary named Susan Franck. Susan has done a lot of excellent work, but what struck me the most was the time and effort she’s put into verifying all the “stylized facts” that people believe about investor-state arbitration. A stylized fact is a fact assumed to be true for the purposes of argument. Economists love ‘em. So do politicians. (Frex, the sylized fact that most U.S. federal spending fits the normal conversional definitions of wasteful.) But stylized facts aren’t always actual facts.

One big stylized fact about investor-state arbitration is that governments always lose. It isn’t hard to see where that comes from. By most reasonable standards, international investment law is an ass. Companies can freely renegotiate contracts with governments, but governments cannot renegotiate with companies. There are few outs for contingencies short of foreign invasion, and maybe not even then. On the other side, it isn’t hard to believe that most of the time the governments are wrong (like, you know, these guys) and deserve to lose.

Only they don’t lose most of the time. Susan found that governments won 58% of cases. Investors won 38%, and the remainder were settled. (She also found that the average claim was for $343 million, whereas the average award came to only $10.4 million. Go figure.)

So how’s Argentina doing? Well, not too badly. Of its post-2001 cases, 33 have been decided. Argentina won eight, or 24%. Of those eight, it won five in arbitration. Two were annulled and one was overturned by a U.S. district court. It lost nine (27%) ... but annulment hearings are pending in five of those nine. If you put those five into the net-yet-decided category, Argentina’s win percentage goes up to 32%, against a lose percentage of 16%.

Of course, 32% + 16% = 48%. Argentina settled a remarkable 52% of its post-2001 cases! Of these 16 settlements, five (31%) were in electricity, three (20%) in natural gas production and distribution, two (13%) in water and sewer systems, and an additional two (13%) in telecoms. The final one was the Unisys contract mentioned in an earlier post.

The settlements aren’t public,  but we do know a little, thanks to Wikileaks. AES owned several power plants, and dropped its suit once the Argentine federal Congress and the Buenos Aires provincial legislature ratified new rate agreements. Pan-American Energy (a subsidiary of BP) sued at ICSID over the pesification of domestic hydrocarbon contracts and the imposition of an oil-export tax. It dropped its claim after Chubut province agreed to renegotiate its E&D concession, presumably by enough to compensate it for the federal takings. If I may take the risk of presenting my own stylized fact, the other settlements were likely similar: rate hikes or other informal compensation, in exchange for a suspension of the international suits.

The bottom line is that Argentina has not done too badly at ICSID and other arbitrations. Its wins outnumber losses two-to-one, and it settles half its cases, as far as we know without direct compensation. What this implies about the unwillingness of the Fernández administration to pay up in its five unequivocal losses, including one pre-2001 case (for a total liability of $427 million) is left as an exercise for the readers.

Internal devaluation does not work very well

So I read in the New York Times that Greek private sector salaries fell a punishing 22.5% in 2011 in real terms.  That stunned me. So I went to Eurostat. Their figures for nominal labor costs were slightly less impressive, but still shocking. (2008 = 100.)

Internal devaluation at work! Sure, Greece is still working off earlier excessive wage gains, but that should presage an export boom, right? Cheaper workers, cheaper vacations for foreigners, more tourism, more exports.

Except ... well, according to OECD data, output is declining along with wages. (See below; 2005 = 100.) Unit labor costs are down, but not when seasonally adjusted, and really not relative to Germany or the rest of the Eurozone. Maybe that’s not relevant. After all, as long as output per worker in potential export sectors isn’t falling, then economy-wide unit labor costs should not matter ... but only for those industries. They will matter a lot in other industries, which are seeing demand (foreign and domestic, as proxied by the value of output) drop as fast as their wage bill.

Add to that (1) rising debt burdens for Greek workers, since debts aren’t indexed to wages; (2) rising taxes on Greek workers; (3) people’s desire to save in euros (outside Greece or in cash) in the face of a possible Grexit; and (4) the awful awful feeling that comes with people as they watch their nominal incomes plummet.

Put it together, and it isn’t clear to me that the rather surprising unstickiness of Greek wages is going to produce much of a quick recovery even if wages fall substantially more. (There have been a lot of reforms to Greek labor law in the last few years.) Of course, the Argentine experience implies that a Grexit won’t be better ... Argentine GDP plunged 15% in the year after the devaluation, and recovery took place in the context of strong external demand.

I am not optimistic.

Postscript: see “Is Increased Price Flexibility Destabilizing?” by Brad Delong and Larry Summers. (Also found here.) See also “The Barber of Buenos Aires.”

It is not unreasonable to cry for Argentina in most of its cases at ICSID

Bernard Guerrero asks: “You mentioned in an earlier post that Argentina had justice on its side with regards to the various ICSID cases. How so?”

An excellent question! Short answer: most of Argentina’s cases are a direct result of the 2001 devaluation. The Argentine government acted prudently and responsibly in the aftermath of that economic collapse. In the face of 1.2 million job losses, and riots in the streets, it was entirely reasonable to freeze rates, cancel contracts, default on debt, and pesify dollar-denominated bank accounts. (The scale of the turn-of-the-century Argentine depression is immense.) Such actions were legal under Argentine law, and would have been legal in the United States. (FDR confiscated gold back in the day.) Unfortunately, the Washington Convention and associated bilateral investment treaties are more restrictive, which gave foreign companies operating in Argentina the right to sue for damages from the price freezes.

In other words, until Repsol, none of the Argentine cases at ICSID involved straight-up opportunistic property seizures without fair compensation. Rather, they involved emergency measures taken in the midst of a Depression-scale crisis. The companies could not have possibly expected the real value of their tariffs to quadruple in the middle of a 20% crash in GDP and 10% plunge in employment, which is what sticking to the original contracts would have meant. Argentine law sees things that way. American law sees things that way. European Union law sees things that way. But international law, as codified in the Washington Convention of 1965, does not. That is unjust.

Longer answer: Of Argentina’s 46 post-2001 arbitration cases, 32 are a direct result of the 2001 devaluation. If you add in the three cases brought by bondholders, that number rises to 35. Most of the remainder are contract disputes, which also come out of the turn-of-the-century Argentine depression: frex, a cancelled contract with Unisys to modernize the judiciary’s IT systems, or an abandoned project with Impregilo S.p.A. to build a highway. 

It is not hard to find the texts (or at least decent summaries) of the ICSID decisions. Latin Arbitration Law is a good source. UNCTAD also maintains a good database. The Investment Treaty Arbitration site has a list of cases, with all intermediate and annulment decisions. Finally, an easy-to-use list with summaries of the outcome can be found at the IIAPP website. I also recommend the Kluwer Arbitration Blog and (if you can get access) the Investment Arbitration Reporter.

All of the Argentine cases reject expropriation claims, but provide compensation for lost revenues due to contract violations. You can see the frustration by the Argentine lawyers when they argue that expropriation law should not apply in these cases. In fact, most of the arbitral decisions have a very sympathetic tone towards the Argentine arguments. But the law is the law, or at least what the arbitrators say that the law is, and that is that.

To be fair, Argentina has not done badly at ICSID. (If there is interest, I will put up another post with some data about that.) But it is understandable that the government feels that most of the post-2001 cases should not have been brought at all. This is not to support the YPF nationalization, or to claim that Argentina is run by a saintly government interested only in the public good. (To give one egregious example, Argentina settled the Unisys claim for $8 million in April 2007 ... but made no move to actually pay.)

It is to say that there is something wrong with the international system. It was designed to protect natural resource investors against predatory states without the need to call on their home governments. It is becoming a system of unlimited liability which essentially locks governments in to the letter of their initial contracts in a rigid way with no parallel in any domestic legal system on the planet.

In fact, the standards have been getting stricter, to the point where governments are required to grant a “guarantee of stability in a secure environment, both physical, commercial and legal” that may, for some countries, be impossible ... and is certainly more protection than that granted to domestic investors.

Practical upshot? Hard to say ... but the illegitimacy of most of the claims against Argentina will make life much harder for companies with a decent case in a straight-up expropriation. (You can guess which firm I might be talking about.) That is not, I think, a good thing ... but I may be wrong.

Maybe the game isn’t on after all, plus health policy!

Well. There may not yet be an equivalent of Fivethirtyeight for the Mexican election, but the poll of polls doesn’t look good for AMLO. You gotta stare really hard to see any sign of movement. Enrique Peña Nieto (EPN) is still cruising towards victory with a plurality of the vote. (Mexico, I have to add, does not have a French-style runoff system nor an Australian-style transferable vote. This is a bad thing.) 

It seems odd — but not surprising! — that the biggest issues in the election are the least discussed. Oh, they both get mentioned in short ads (see below) but not much in concrete terms. I am referring, of course, to crime and health policy. The latter should be a big issue. While Mexico has made great strides in public health over the past 18 years, it is still far from universal coverage. So what do the two main candidates propose? (My apologies, Josefina.)

Well, both AMLO and EPN think it ridiculous that Mexico has a multiplicity of public health systems: one for formal sector workers, one for government employees (both run like the U.K. National Health Service) and a bare-bones People’s Insurance program for the poor. So it will be no surprise that both are calling for universal coverage. AMLO, however, wants change to be incremental. He wants to allow patients to receive services in the installations of any of the public health programs, a de facto conversion of the federal government’s role to something more like a single-payer system. The problem, of course, is a combination of complexity and vested interests. Peña Nieto is more direct: he wants a new Universal Social Security System, which he claims would cut overhead expenses from 9.5% to 3% and de jure move Mexico to Canadian-style Medicare. He also wants to provide more free medicines via a voucher system, with no word on how to pay for it.

So who would you trust? EPN has a better, more-concrete plan. AMLO is more serious about these things, and incremental change is generally easier. At the end of the day, AMLO is more honest about the inevitability of the need to throw more money at the problem ... but at the end of the day, a President AMLO is also more likely to be hamstrung by Congress.

Edge: AMLO. Not that it matters.

Game on, México!

 Wow. Oh wow. Big news in the Mexican presidential race.

Courtesy of Ganchoblog, it seems that the Mexican presidential blowout that we all expected for Enrique Peña Nieto may not actually happen. The Reforma newspaper runs reputable polls, and they show a sudden collapse in PRI presidential support from 42% to 38%, with AMLO rising from 27% to 34%. Now, I remember drinking pulque (horrible stuff, by the way) with some friends in Melchor Ocampo (awful town, I should add, although worth visiting) and being told that the polls meant nothing, AMLO could pull it out. I scoffed. Distant relatives in Huixquilucan (a very much not-awful town, at least if you like bad science fiction movies from the 1970s, but really not worth visiting unless you know somebody who lives there) said the same. But they were PRD partisans, and so I scoffed some more.

But they were all correct! If this poll holds, then the electoral game is on. We will have a real race on our hands, and not a PRI blowout. Both Melchor Ocampo and Huixquilucan are in Mexico State and under the Peña governorship. I think that gave my friends tacit knowledge about the fragility of his support that nerdy old gringo me could not appreciate. I think Mr. Hootsen, who also lives in Mexico State, would concur.

Were I voting in the election, a poll like this would pose a dilemma. I despise AMLO for his actions after the 2006 election. But I suspect, for reasons I find distressingly hard to verbalize, that he would be less damaging to Mexico than a President Peña. Which means that strategically speaking, if AMLO had a chance, I would want to cast a strategic vote for him. Except ... his behavior and statements in 2006. The questions raised by the commentators here are becoming much more valid. Assuming that the poll holds and we really do have a race, for whom should a concerned Mexican citizen vote?

I would like to end with a call for someone to start a blog like Fivethirtyeight, aimed at the Mexican elections. So far, the Reforma poll is an outlier. We have no way of judging its accuracy or predicting the result. Mexico could use a Fivethirtyeight. I nominate Diego Valle to start it.