Friday, February 28, 2014

Flexible Exchange Rates, MMT, and the Ruble Crises of MMT

Ramanan continues on his MMT critique focusing on Forex. sovereignty, and flexible exchange rates. The standard MMT position around "currency sovereignty" is simply that a state has the ability to issue its own currency and thus is "sovereign" in it's matter. State obligations, in its own currency, are easily extinguished by it issuing more, and moreover, were initially funded by it issueng currency.

A simple thought experiment: if I was to establish winterspeak-land tomorrow, no one could pay a tax denominated in winterspeak-bucks because none would exist, nor could anyone buy a winterspeak-bond because they would have no winterspeak-bucks to do so with. The original winterspeak-buck would need to come from somewhere, and that somewhere would be me (else it would be counterfeit).

This is in some ways the most intuitive part of MMT, and yet it seems to cause problems when you have higher level discussions on more technical points. Regardless, let's turn to the real world and see just what the limits are to currency sovereignity, what the real world policy space is re: floating exchange rates, and what a countries' Forex obligations actually are.

First, let's establish that any goods and services that a state can generate and provider for itself can be entirely managed in its own currency. However, most states need some imports, and those imports may or may not be available in the state's own currency. In those cases, the state will need to either export real goods and services to get the required currency, or will need to be able to swap it's currency for the target currency on a forex market.

In the currency swap scenario, the state can either let the currency float freely, or it can attach it to some sort of peg. If the currency floats freely, then currency sovereignty it maintained. If there's a peg, then sovereignty is compromised as the state will be obligated to maintain that peg, or default.

In practice, most countries operate on what I called a "mixed currency regime" where the citizens use a mix of local and foreign currencies. This is certainly true in third world countries, and this mixed regime reflects how sovereignty in those countries is also weak -- there is usually limited law enforcement, corruption, weak property rights, etc. As the sovereignty of the state increases, you find mixed currency regimes becoming less mixed, and if the state is hyper-sovereign, ie. has the ability to dictate, or at least strongly influence the affairs of other states, then you find its currency beginning to dominate in those other countries as well. The US$ is exhibit A.

Ramanan writes:
Now, it is clear from the above quote that Mitchell admits that nations with government have a constraint on fiscal policy. But the more troublesome fact is that he presents it as if the government doing this had some full volition to not have been indebted in foreign currency.
It depends on how much the country wants to import. Free trade has costs instead of benefits, and a country can easily develop internal auto industries, for exmaple, if it wants to. The cars won't be as good as if they came from Japan, but it could do it. Aside from certain commodities, countries do have the option to trade less, and this become less indebted in a foreign currency, although that also has costs.

He continues:
So the Neochartalist story that somehow the government shouldn’t borrow in foreign currencies is vacuous. Only a few nations have the ability to attract investors to purchase their debt in domestic currency and typically these nations are successful in international trade. But this by no means guarantees continued success – if net indebtedness to foreigners keeps rising relative to output because trade in international markets for goods and services turns weak, then output gives in. And a shift in investors’ portfolio preferences also can lead to the original sin, even if one starts with zero government debt in foreign currency.
I don't agree that the MMT story is vacuous, I think it highlights another real cost and risk to free trade. He ends on what I consider another non-sequitor:
Now the backfire effect: in the “modern monetary theory” blogs, examples such as those of Pakistan are presented as if it was Pakistan’s policy makers huge error to have borrowed in foreign currency and to their fans this appears to strengthen the view that in the supposed world which the Neochartalists fantasize, there is no balance-of-payments constraint. And the error is the failure to recognize that “money” has an international aspect in addition to what it has to do with the government and banks.
At present, the solution is for the world leaders to provide a coordinated fiscal expansion and induce the creditor nations to increase domestic demand and hence increase latter’s level of imports. But the long term solution is to move away from a system of free trade. And that is far from the “MMT” overkill description of the world and overly simplified solutions.
I'm very familiar with Pakistan, and borrowing in dollars with a peg certainly constrains Pakistan's ability to use fiscal policy to boost domestic demand if that's what necessary. Honestly though, Pakistan's bigger problem is it's inability to generate savings demand for rupees from it's domestic population because it has very weak tax collection. Stronger tax collection would mean a bigger sink for rupees, reducing inflationary pressure, and letting it run higher deficits. Pakistan's need to borrow in foreign currency stems directly from it being unable to tax in it's own, although I do not want to imply that if it could tax in rupees that would fund it's ability to spend in rupees. The mechanism isn't a causal funding mechanism. The Pakistani government is a limited sovereign.

The ruble example of 98 is interesting because there, Russia defaulted on ruble denominated debt, which it could theoretically have honored. The contagion mechanism there was actually first world hedge funds, who were forced to sell other positions because their foreign debt portfolios were being squeezed.


One of the more interesting MMT insights, as stated by Mosler, is "exports are a cost, imports are a benefit", which is the opposite of the usual narrative where export driven economies, like Japan, are hailed while import driven economies, like the US, are said to be more vulnerable. When you look at straight goods and consumption, it's obvious that a country wants goods (which are hard to produce) more than it wants it's own currency (which is trivial, the trick is in getting it accepted). There are other reasons why an import heavy economy generates it's own costs, but if you had to choose between beads or the island of Manhattan, I hope the choice is clear even if you aren't conversant in MMT.

Wednesday, February 26, 2014

The future of news

Marc Andreessen has a post on the future on news, and it's interesting to see how his experience has colored his views (as happens to us all). Some interesting elements:
The news business is a business like any business. It can and should be analyzed and run like a business. Thinking of news as a business is not only NOT bad for quality, objective journalism, but is PRO quality, objective journalism. A healthy business is the foundation for being able to build high quality products, and to do so sustainably. That includes journalism.
The main change is that news businesses from 1946-2005 were mostly monopolies and oligopolies. Now they aren’t. The monopoly/oligopoly structure of newspapers, magazines, and broadcast TV news pre-‘05 meant restricted choice and overly high prices. In other words, the key to the old businesses was control of distribution, way more than anyone ever wanted to admit. That’s wonderful while it lasts, but wrenching when that control goes away.
Really?

The news "business" was split into two parts -- tabloid journalism (or the old "yellow press") who dealt with sensationalist headlines, and the "serious" press who felt more responsibility, for good or will, of the individuals they were trying to influence. Within both of these, there was the business side, focused on advertising and classifieds, and the content itself which was run quite separately. The Church and State divide should illustrate how newspapers were never really, or at least not entirely, in the business of selling information for money. Nor is it clear that there is any market at all for the traditional, high minded, "serious press". Maybe the NYTimes should be run as a non-profit where they can ditch all the ads and focus on telling us what to think.

Some other titbits:
Right now everyone is obsessed with slumping prices, but ultimately, the most important dynamic is No. 3 – increasing volume. Here’s why: Market size equals destiny. The big opportunity for the news industry in the next five to 10 years is to increase its market size 100x AND drop prices 10X. Become larger and much more important in the process.
"Market size equals destiny" is a great sound bite, and I see how Marc's experience with Netscape, as the web exploded, and now as a VC who listens to pitches all day, would shape this idea. There is also a great deal of truth to it.
An obvious one is the bloated cost structure left over from the news industry’s monopoly/oligopoly days. Nobody promised every news outfit a shiny headquarters tower, big expense accounts, and lots of secretaries!

Unions and pensions are another holdover. Both were useful once, but now impose a structural rigidity in a rapidly changing environment. They make it hard to respond to a changing financial environment and to nimbler competition. The better model for incentivizing employees is sharing equity in the company.
Again, really? I don't think Unions and pensions were about motivating employees, I think they were about giving labor more bargaining power vis a vis management so labor could get more profits for themselves. And while Marc's had great luck with equity, I'm sure he knows from his VC position how that just isn't true for the median employee across his portfolio companies.

I'm not arguing for either unions or pensions here. Both have serious problems. However, in this current climate where Silicon Valley billionaires are being criticized for being out of touch, "let them eat equity" may not be the most appropriate thing to say.


Friday, February 21, 2014

Facebook lost the social graph

Various companies have tried to build out identity systems on the web, and none of them were successful until Facebook came along with it's social graph. It did this by offering a service where real world identity was key to it working (as LinkedIn did before) but then crucially, offering that as a platform via FBConnect (which Linkedin did not do, and maybe it could not have given it's more narrow focus).

But on phones, such social graphs already exist and they are called your address book.
With new consumer Internet companies being started every day — and online consumers notoriously fickle — a next big thing will always be rising up to threaten Facebook’s dominance. As a result, Facebook, like other technology companies, is left playing an expensive game of Whac-a-Mole, scrambling to buy its newest competitors and keep those users out of the hands of its rivals.
“This exposes the strategic fallacy behind Facebook, which was the idea that there was going to be a monopoly on the social graph, and that Facebook was going to own it,” said Keith Rabois, a partner at venture capital firm Khosla Ventures. “That’s not true, and I don’t believe Facebook will constantly be able to buy its way out of this structural challenge.”
Whatsapp does not have a social graph, it leverages the social graph in every address book. And it was only successful because of how operators chose to charge for text messaging instead of offering it for free, or as a more reasonably priced option. The inability of telecoms to profit off mobile is remarkable.

update: Albert Wenger agrees and thinks that Facebook overpaid for WhatsApp:
Having had some time to think about that I am now convinced that this deal makes no sense.
Why? Because phone number based messenger apps can bootstrap very rapidly off the graph that is contained in people’s address books. We are witnessing that now with the Telegram Messenger app which apparently signed up nearly 5 million users yesterday. The UIs of all of these apps are virtually identical and are also extremely similar to the basic SMS UI that everyone around the world knows and understands. The combination means there is virtually no enduser lock in at the messaging layer.
I wonder to what degree this was motivated by Mark's own history? His experience has consisted entirely of "make something kids use to connect with each other and become wealthy". Facebook gained traction right away and has not had to deal with any real failure. It's also not how people connect on their phone.

Wednesday, February 19, 2014

Palley on MMT

Ramanan is very impressed by Palley's MMT critique (pdf) and while I think there are good MMT critiques out there, I don't think Palley's is one. I also think, crucially, he missed why MMT has (what little) appeal that it does.

Palley is correct, though, in pointing out that a failure to phrase things in terms of DGSE means that the theory has not made headway in the academy. I know JKH hates the term "paradigm", but it was used by Kuhn in a particular way for a particular reason which is applicable here, and Palley lays it out very clearly in his introduction:
This question can only be answered by placing that power within a theoretical model and exploring its implications. For the last seventy years the language of macroeconomics has been small scale simultaneous equation models with dynamic adjustment mechanisms attached to explore issues of stability. Proponents of MMT have a professional obligation to provide such a model to help understand and assess the logic and originality of their claims.
I am not opposed to models at all, as I've stated repeatedly on this blog, but I also know that if you start with radically different assumptions you're not going to make progress and MMT attacks orthodox macro at the assumption level. Moreover, models are built within pre-existing interpretations, you do not write out a bunch of equations and then do what they say, and when you're changing the interpretation, beginning with equations may not help.

Before diving into the Palley paper more deeply, let me also say that the only reason MMT currently has it's current moment is because orthodox macro has failed. It's been failing in Japan for 30 years, but now that we have the same situation in the US, and we have the internet and write in English, the failure has become pretty clear. Palley is right in asking MMT to explain price stability and full employment, provide a credible theory of inflation, and justify monetary policy and it's effects; but frankly I think all of those questions can be turned on textbook macro as fairly, and with greater urgency since they are the ones in the drivers seat. Because right now there's a big gap between what's between the book jacket and what's outside the window.

Palley spends some time talking about how it is common knowledge that "sovereign issuers of money are not constrained financially in the normal sense". Great. And yet we hear continual talk about the US Govt running out of money. So it may be common knowledge for Palley, but it is not common knowledge more broadly, and if MMT is working to popularize that understanding he should be applauding it not yawning. A little support towards the common good, please.

Oddly, Palley makes no distinction between the US situation, where the deficit is in US$, and the EU situation where member country deficits are in euros. So there seems to be a real difference between the Palley view and the MMT view on this.

Like Palley, I am stunned by the claim that MMT rejects counter-cyclical fiscal policy. Certainly contrary to what I've heard, and if Wray really does take this position then we'd have to disagree. There's a longer post I keep meaning to write about "employment" where I want to talk about this further. However, I don't agree that MMT has no theory of inflation, nor do I think there is a single "theory of inflation" as the observed behavior can have different causes. The Phillips curve has, at times, not accorded with reality.

If I'm to engage with the model Palley sketched out (thanks!) it would be to point out several areas where his sketch conflicts with MMT.

1) Palley asserts that the deficit (D) be zero in a static economy, thus requiring T - G = D = 0. However, in MMT aggregate demand is conditional on the need for what I'll call "savings" for now to be met, "savings" defined as G-T. And this demand may be non-zero. It is, in fact, the very variability of this which Palley does not model, and so misses the core MMT contribution.

2) Palley also sets up a condition where Gmin cannot exceed Tmax. Again, in MMT, why? Certainly if the policy interest rate is set to zero, any difference between Gmin and Tmax can be made up by the consolidated fiscal and monetary authority (which Palley supports earlier) borrowing from itself, or running in overdraft conditions, depending on how you want to think about it.

3) And finally, I would ask that Palley himself puts forward a theory of inflation in his model beyond the footnote where he says "to avoid inflation the high-powered money stock must grow at the rate of growth". Why? What exactly is the automatic transmission mechanism between high-powered money (a stock) and inflation (necessarily a flow phenomenon)? There is no automatic mechanism, on a volitional one (spending) and it is here that MMT adds something beyond standard Keynesianism.

Finally, I too have concerns with ELR, but I don't believe they are part of MMT, merely a policy recommendation, and my concerns are different from Palley's, who attacks it from the Left.

To come: flexible exchange rates where I give first hand testimony from the great ruble crises of '98

Friday, February 07, 2014

Bitcoin Mania

A surprisingly cogent article from Fortune, of all places, on Bitcoin:
Bitcoin's primary significance is not about whether it supplants cash. It's about a revolutionary computer-science breakthrough that has the potential to upend all sorts of established industries.
It's rare that a pure technology comes along, with so much power, but so little obvious application.