Comments Guidelines

All comments are pre-moderated. No spam, slurs, personal attacks, or foul language will be allowed.

Monday, June 29, 2015

Will the U.S. keep winning indefinitely? ISDS, that is

Now that Congress has given the President fast-track Trade Promotion Authority, the first agreement to be considered under these rules (no amendments allowed, up or down vote in 90 days) will be the Trans-Pacific Partnership (TPP). As you know from previous columns, one of the most worrying aspects of the TPP is its expansion of investor-state dispute settlement (ISDS), wherein private firms can bring their disputes with governments not to courts, but to international arbitration (usually through units of the World Bank or the United Nations), where legal precedent doesn't matter and appeal is all but non-existent. Moreover, as the Consumers Union has long argued (recent example here), arbitration has a well-known pro-business bias. That's why so many of your agreements with cable TV providers, financial services companies, and many more have fine print requiring mandatory arbitration, keeping you from getting your day in court if something goes wrong.

The response from the U.S. Trade Representative's (USTR) office has been, "Not to worry! The United States has never lost an ISDS case." The linked document goes on to claim that worldwide, only 1/4 of corporate plaintiffs have won cases against governments. But a new analysis by the International Institute for Sustainable Development (IISD),* using the same data source the USTR cites, comes to a very different conclusion based on its most recent update, the 2015 World Investment Report from the United Nations Conference on Trade and Development (UNCTAD). Moreover, we can see that countries with even more trustworthy court systems than that in the U.S. have lost ISDS cases. The Rule of Law Project, an initiative of the American Bar Association, has ranked 102 countries on the administration of justice and freedom from corruption, and puts the United States at #19 with a score of 0.73. Yet #14 Canada (0.78) has already lost ISDS cases, and both Canada and #10 Australia (0.80) are currently on the hook for major new cases (Eli Lilly and Philip Morris, respectively), that would overrule decisions by the countries' respective Supreme Courts. So, even if governments have only lost 25% of ISDS cases, it's unlikely U.S. luck will hold out indefinitely, if countries with better court systems are losing.

But it's worse than that. UNCTAD's database of known ISDS cases and their outcomes shows that in all cases decided through the end of 2014, the investor won 27% of the cases compared to 36% won by the state (see Figure III.10, p. 116). But another 26% of the cases are listed as "settled," which often (but not always) means the respondent agrees to make some payment to the plaintiff to keep the case from going to arbitration. Public Citizen has a list of ISDS cases under prior U.S. trade agreements with examples of settlements that do and do not contain payments (see, for instance, NAFTA cases against Canada).

Moreover, as IISD attorney Howard Mann argues, if we separate out cases between jurisdictional determinations and determinations on the merits of the case, things look even worse for states. While only 71 of 255 cases (this excludes the "settled" cases) were concluded by a decision of the tribunal having no jurisdiction, Mann points out that all 255 cases effectively had decisions on jurisdiction, i.e., cases with final decisions had to have rulings that the arbitrators had jurisdiction. In that case, Mann says, "Investors, therefore, have won 72 per cent [184/255] of jurisdictional determinations." And of the decisions on the merits of the cases, investors won 111, or 60%, of the remaining 184 cases. This calculation suggests that states are losing ISDS disputes at a much higher rate than normally portrayed. As if that's not bad enough, the new World Investment Report finds that in 2014, of the 15 ISDS cases decided on their merits, states lost 10 (2/3) of them. In 2013, it was even worse for states, with investors winning 7 of the 8 cases decided that year (p. 126). If these higher proportions continue, obviously the proportion of investor victories will increase beyond the current 60% total.

Bottom line: The threat to regulation, democracy, and the rule of law posed by investor-state dispute settlement is very real. The U.S. Trade Rep's  reassurances that the U.S. has never lost in ISDS don't even make it likely that will continue into the future. We need to pressure Congress to vote down the TPP when negotiations conclude.


* Important disclosure: I have consulted for IISD several times since 2007 on investment incentive issues.

Cross-posted at Angry Bear.

Saturday, June 13, 2015

Elon Musk has received billions in subsidies

While receiving subsidies is nothing new for the Forbes 400 or even multi-hundred millionaire pikers like Mitt Romney, a recent story in the Los Angeles Times (via Good Jobs First) shows that Elon Musk (#34 in the Forbes 400) is a champion at getting subsidies for his companies. According to the Times article, Musk's three companies, Tesla, Solar City, and SpaceX, have received a total of $4.9 billion (nominal value) in subsidies over the years. The article says that Tesla and Solar City stand out in the importance of the subsidies relative to the size of the company.

While SpaceX has received only $20 million, both Tesla and Solar City have received over $2 billion each, if you count the value of the subsidies their customers have received for buying Tesla vehicles and Solar City installations. This is more significant in the case of Solar City (about $1 billion) than for Tesla (about $321 million). Even without these sums, the companies have directly received about $3.5 billion, most notably for the new Gigafactory in Nevada and for a solar panel facility in Buffalo, New York.

Regular readers will remember that I have long argued in my books and elsewhere that these subsidies represent a transfer from average taxpayers to the much wealthier owners of the companies involved, worsening the already substantial inequality in the United States. These investment incentives have to be offset by higher taxes on others, reduced government services, or higher levels of government debt. While they are not the biggest driver of inequality, they do their part. Moreover, location subsidies reduce the country's economic efficiency: It may well have made more economic sense to locate the battery Gigafactory as close as possible to Tesla's assembly plant in Fremont, California.

While Musk refused to be interviewed for the Times story, he responded the next day on CNBC. Among other things, he argued that it was wrong to report a single figure for subsidies, which makes it seem like he received one big check. This is right as far as it goes. However, I think it would make more sense to give a single present value for the subsidies rather than the nominal value, which overstates the value of multi-year subsidies such as those for Tesla. Moreover, as Good Jobs First points out, it is perfectly necessary for taxpayers to know what their long-term liabilities are for multi-year subsidies in order to properly assess the impact on government finances.

Musk also defended the Tesla subsidies as merely necessary to make the project happen faster, rather than necessary to happen at all. Yet it conducted a multi-state auction in an all-too-common use of its location decision for rent-seeking. As I analyzed at the time, the deal was below average in terms of cost per job and aid intensity compared to other automobile facilities, and it is 13 times larger than Nevada's previously largest incentive package.

Ultimately, the Musk story is far too familiar on a number of dimensions. Most importantly, it is a tale of rent-seeking and the policy/political drivers of inequality.

Cross-posted at Angry Bear.

Friday, June 5, 2015

GE threatens to leave Connecticut UPDATED

A non-blogging friend has brought to my attention the fact that General Electric is threatening to move its corporate headquarters out of Connecticut in response to proposed tax increases in the state budget. The company, based in Fairfield, objects to increases in the taxes for data processing and corporate headquarters. Insurance companies Aetna and Travelers have also issued similar threats.

In GE's case in particular, this is pretty rich. The New York Times reports that GE made $14.2 billion in 2010 and received a federal tax refund of $3 billion. It is a company that touts bringing manufacturing jobs back from China but conveniently omits mentioning that the new jobs, in Louisville, pay $13/hour rather than the $22/hour they paid before they left for China. A model of corporate virtue it ain't, yet President Obama sees fit to lean on chief executive officer Jeffrey Immelt as one of his top business advisers.

Regular readers no doubt recall that every time a threat like this is made, vultures start to swoop in to attract the potential relocator to their state with a long list of goodies. So it should come as no surprise that a mere three days after GE first floated this idea, Tampa, Florida, has put GE in in its sights. We've seen this story many times before: Boeing and Sears immediately spring to mind. As always, the possibility of receiving relocation subsidies makes relocation less expensive and makes it more likely that a company's current home will have to give concessions to make it stay. Job piracy and job blackmail are intimately related.

As my friend points out, it's not surprising that Connecticut is running a big budget deficit. In just the past few years, according to the Good Jobs First Megadeals database (March 2015 spreadsheet update), the state gave $313.75 million to Schupp & Grochmal (2007), $89.5 million to Starwood Hotels (2009), $291 million to Jackson Laboratory (2011), $115 million to Bridgewater Associates (2012), and a whopping $400 million to United Technologies (2014). This last deal is fully 20% of the entire 2-year budget deficit facing the state, $2 billion.

A showdown is looming in this newest case of raw corporate power. Yesterday, the Connecticut legislature passed the budget, though leaders signaled a willingness to consider small changes when it comes back for a special session. The same day, however, Immelt emailed employees to let them know a task force had been set up to look into relocation.

Stay tuned!

Update: More evidence that GE is a whiny, hyper-aggressive tax avoider:  http://www.courant.com/opinion/op-ed/hc-op-cibes-ct-business-taxes-not-high-0609-20150608-story.html

Friday, May 22, 2015

May Tax-Cast is Out!

The May Tax-Cast from the Tax Justice Network is just out. Highlights include tax transparency from politicians in Pakistan and around the world, as well as an analysis of the recent UK general election.

Tuesday, May 12, 2015

TPP blocked, at least for now

Breaking news: The United States Senate failed to end a filibuster on giving the President fast-track negotiating authority for the Trans-Pacific Partnership (TPP), the Transatlantic Trade and Investment Partnership (TTIP), and any future deals for a six-year period.

According to The New York Times, the bill failed with only a 52-45 majority for it, when 60 votes were needed to end the filibuster. The biggest complaint among swing Democrats was that the TPP does not have enforceable provisions against currency manipulation -- such as practiced by China which, though not currently involved in the negotiations, would seem like a logical future party to the agreement. The Times reported that Japan and Malaysia are both opposed to this provision.

For the time being, then, we are spared an expansion of investor-state dispute settlement and further unnecessarily strong protections for intellectual property (patents, trademarks, copyright, etc.). Unfortunately, if the proponents of the measure can reach a compromise with a group of eight Democrats (including Ron Wyden of Oregon) on enforcement measures, then they would have the 60 votes they need.

Monday, May 11, 2015

Why subsidize data centers?

A number of authors (Good Jobs First, David Cay Johnston, me, and me, among others) have pointed out that data centers (aka server farms) in the United States create very few jobs, yet receive state and local government subsidies that routinely exceed $1 million per job. I'm sure you already know that numbers like those make me ill: the typical automobile assembly plant will receive $150,000 or so per job, and require all sorts of component facilities to feed it -- though, sadly, economic development officials often given incentives to the supplier plants as well.

So why $1 million or more per job? Data centers pay reasonably well, and the biggest are connected to famous tech names like Apple, Google, and Facebook, but it seems to me that it's hard to get around the facts that there just aren't that many jobs, and they don't require an army of supplier facilities bringing indirect jobs.

But surely the competition for jobs is so steep that governments have no choice but to subsidize them? Actually, no. Aside from the fact that $1 million per job probably gives away more than the value of the investment to the government, my investigations have turned up multiple examples of companies building data centers without incentives.

One I've mentioned here before: American Express in 2010 built a $400 million data center in Greensboro, North Carolina, without any incentives at all. The leading explanation has been that Amex had already decided it was going to close a 1900-job call center in Greensboro (announced in 2011), a move it knew would trigger clawbacks of any incentives on the 50-150 job data center -- so it didn't bother seeking subsidies. Did I mention that North Carolina has cheap electricity?

More recently, I have found four Google data centers that opened or expanded without incentives in the last few years. New and expanded facilities in the Netherlands, Ireland, Finland, and Belgium all take advantage of cooler temperatures to reduce their electricity use. While Google did not respond to my email asking whether it received subsidies for those facilities, and IDA Ireland similarly was unresponsive, the Netherlands Foreign Investment Agency did respond with a confirmation that it had provided no "state aid" (EU-speak for subsidies) to the brand-new $773 million, 150-job data center opening in Groningen province in 2017. In addition, a search of the EU's Competition Directorate case database did not reveal any Google state aid cases for data centers. Thus, it appears that none of these cases received incentives.

So why did Google demand over $140 million (present value) in subsidies from North Carolina back in 2007? I think we're looking at the "usual suspect" once again, rent-seeking. Of course, North Carolina couldn't foresee the Amex no-incentive deal that didn't happen until 2010, but now that we can see how Google and American Express do business when they have to, it's time economic development officials around the country learned to "just say no" on data centers.

Cross-posted at Angry Bear.

Thursday, April 23, 2015

Moving the goal posts on ACA success UPDATED

Right. So the the same day that I posted about the substantial fall in the uninsured rate for adults that we have seen since Obamacare went into effect, a conservative writes at the Wall Street Journal making exactly the same arguments that Matt Yglesias had refuted. Cliff Asness writes in the WSJ:
That more people would be insured was never in dispute. If you mandate that people buy something, penalize them if they don’t and give it away to some, more people will end up with it. The proper response to this is: Duh.
So, as I said, Yglesias had already refuted this, giving a number of examples of conservatives who predicted there would be no reduction in the number of uninsured Americans. Today, Paul Krugman takes us to Jonathan Chait's response to Asness, where of course he piles on more examples of conservatives predicting a failure to improve the uninsured rate. Then he goes further. Asness wrote that a critical issue was "how many people covered by ObamaCare were previously uninsured." You can probably guess Chait's answer: "Well, that’s why you measure the net number of uninsured people, not just the gross expansion of coverage under Obamacare." Which leads us back to the chart showing the substantial fall in the uninsured rate that was in my last post (and Yglesias', Krugman's and no doubt many more besides).

The latest round is that yesterday Asness responded to Chait. Here is where the goal post move comes in:
In contrast the rise in coverage is heralded by a myriad of Obamacare supporters as one of two major pieces of proof the law is working. But, how can something we knew before the fact be proof of anything?
Did you catch that? If we predict that something good will happen as a result of a new law, and that good thing happens, it doesn't count as proof that the law was good. This is silly. We didn't actually know the insurance rate would fall, but we had economic models that told us it would. So not only is the fall evidence that the law is working, it's evidence that the models were right!

Somebody wake these people up.

UPDATE: @HaroldPollack points me to a new J.D. Power survey finding that people who signed up for insurance on the exchanges were more satisfied (696 out of 1000) than people with non-exchange plans, usually through employers (679 out of 1000). People re-enrolling on the exchanges scored even higher, with a score of 744 for people who re-enrolled on the Exchanges. Private plans offering multiple options were able to reach the 696 average for Exchange enrolees, which means that companies offering one insurance option had to be doing substantially worse than 679. Not surprisingly, new enrolees for 2015 were a large 55 points more satisfied than 2014 enrolees, who of course went through the disastrous rollout of healthcare.gov.  So people like their subsidies and they like their actual insurance policies, on average. Maybe that's why Republican Senators are getting antsy that there will be hell to pay if the Supreme Court rules for the plaintiffs in King v. Burwell.

Cross-posted at Angry Bear.