Forget Summers and Yellen: Let a Bot Run the Federal Reserve

1956 - ”Forbidden Planet” - Robbie! by x-ray delta one, on Flickr
Creative Commons Attribution-Share Alike 2.0 Generic License  via  Flickr 

The shameless campaigning and cat fight over who should be our next Federal Reserve Chair has only served to underscore the inevitable human flaws and lack of omniscience of whoever is ultimately appointed, all while further politicizing and debasing the office. I argued four years ago a bot could do that job and feel obligated to make the case again. Other than changing the date, this still makes perfect sense:

Software Bot To Be Nominated Chairman of 
Federal Reserve System

Cutting edge technology tapped to bring stability and
consistency to monetary policy

WASHINGTON D.C. – August 25, 2009 — President Barack Obama today announced he intends to nominate Monet 3.0, a software bot, to become Chairman and a member of the Board of Governors of the Federal Reserve System. As Chairman, Monet will be charged with conducting the nation’s monetary policy by influencing money and credit conditions in the economy.

“Software bots today are successfully outperforming the world’s best human practitioners in complex endeavors like chess, and they do so without irrationality or exuberance,” said President Obama. “Despite the bot’s French-sounding name, I am confident that Monet 3.0’s discipline and transparency will bring price stability and foster the economic growth required for a full economic recovery.”

Monet was born in a lab at Stanford University in the early 1990s and is currently in version 3.0.  The software instantiates a modified version of the Taylor Rule.  Source code for Monet is available for broad inspection and reuse under the modified BSD license at

Monet’s nomination requires approval by the Senate and a bout of hyperinflation that makes the Hungarian episode of 1946 look modest.

I could update the proposal with a dash of Big Data and or integrate with the Bitcoin blockchain (we’ll leave pioneering on that front to Iceland), but I think the original idea still suffices. It is easy to make contemporary arguments for our bot (e.g. even a “headless” bot will be more user friendly than Larry Summers). All the bird-watching speculation about hawks and doves becomes irrelevant, as our bot will mindlessly execute a direct link between objectives and policy. And in these times of government fiscal constraint, beyond electricity and the occasional motherboard upgrade, moving to a bot will be cheaper than a new human occupant.

Making a bot the Fed chair will also help the broader Federal Reserve Board of Governors increase their understanding of important economic dynamics. Members of the Federal Reserve have not always demonstrated the greatest empathy for the consequences of their policy. Seeing a robot take a colleagues’ job firsthand should give them a deeper appreciation for the situation of the American worker.

Plus, the all-knowing NSA is already moving down this path to cut fickle and fallible humans out of the loop and the Federal Reserve, perhaps the only other government entity with comparable reach and influence, would not want to fall behind in the Race Against the Machine (a topic I will get to one of these days).

I will keep posting this every four years until a bot takes its rightful place as Chair of the Federal Reserve.

Press Releases We’d Like to See: Iceland Embraces the Bitcoin Economy

Continuing our series of monetary-themed press releases we’d like to see (but probably won’t):

Iceland Embraces the Bitcoin Economy
Unveils comprehensive program for
21st century financial leadership

REYKJAVIK, Iceland — May 1, 2013 — The Republic of Iceland today announced a comprehensive program to make Iceland the leading nation state in the emerging Bitcoin economy.

“Bitcoin is a math-based, digital currency that provides the foundation for addressing many of the financial challenges afflicting both the real economy and the digital economy today,” said Snorri Thorodinsónssónssónssónssón, Minister of Finance and Economic Affairs for the Republic of Iceland. “With our abundant, low-cost geothermal power and our formative experiences in the recent global financial crisis, no nation is better positioned than Iceland to help develop a new set of digital financial institutions the 21st century so desperately needs.”

Bitcoin’s many benefits – including cryptographically secure transactions, lower transaction costs, fraud mitigation, nearly instantaneous payments across the globe, transparent accounting and reduced dependence on “too big to fail” banks – make it the obvious foundation upon which to build a new financial system. As Iceland develops its digital economy powered by geothermal energy, a Bitcoin-based economy broadly benefits digital businesses hosted in Iceland, creates opportunities for new Bitcoin-based businesses and increases overall economic efficiency.

“Iceland intends to be at the vanguard of the new digital economy,” Thorodinsónssónssónssónssón continued. “While some may put their confidence in the resolve of policymakers, we put our confidence in the cryptographic assurance arising from the second preimage resistance of the SHA-256 hashing algorithm.”

The government’s comprehensive program includes:

  • Bitcoin Payments – Iceland will immediately accept Bitcoin as valid payment for all government transactions including taxes, customs, duties and other fees.
  • Government Finance – Iceland will sell the first Bitcoin-denominated government bonds and offer its citizens the option of Bitcoin-denominated assets in government pension plans.
  • Economic Development – in conjunction with the burgeoning data center industry in Iceland, the government will incentivize new Bitcoin-based businesses, including exchanges, mining operations and a new generation of financial institutions, to locate and grow in Iceland.
  • The Icelandic Bitcoin Furnace – the Icelandic government will sponsor the crowdsourced design of a custom, ASIC-based Bitcoin mining device for the home. The device will offer terahash-class mining performance along with highly efficient heat recirculation for the long Icelandic winters. The government has reserved significant 28nm semiconductor fabrication capacity for production of the final design, which will also be made available as an open source reference design.
  • Advisory Services – Iceland will make its expertise in Bitcoin available to other sovereign nations interested in joining the 21st century financial economy through a range of consultative services. These services will initially be available through Icelandic embassies in Athens, Dublin, Lisbon, Madrid, Nicosia, Paris, Singapore and Zurich.

No diacritical marks were harmed in the making of this post.

And if you are a Bitcoin-related (or other cryptocurrency) startup, whether you are in Iceland or not, I want to talk to you. See the About page for information on how to contact me.

Economic Forecasting in Alternative Universes

<dweeby econo-skepticism follows>

The New York Times Magazine today has an article on the past, present and future of the Obama administration’s economic policy.  It is a good if long read and chock full of inside baseball, political speaking points delivered with some repetition lest you miss them and, most entertaining, anonymous score-settling amongst the recently departed economics team.  The narrative is mostly cheerleading, ex post facto rationalization and blame-shifting for the last two years.  And despite an intent to paint a positive picture going forward, specific points of the story are quite damning:

  • In a dramatic meeting December 16, 2008 before taking office, the new team was “warned the country was in far worse shape than anyone realized.”  Despite this deliberately seeded anecdote, one excuse offered is they didn’t understand the true magnitude of the crisis: “The problem was that the baseline economy was in worse shape than even the grim assessment of that Chicago meeting in late 2008.”
  • Economists from both the left and the right are quoted saying the strategy and guiding principles for recent economic policy are unclear: “This was all new to Obama, who, unlike Bush or Clinton, had never managed even a state economy.”  The lack of strategy is charitably described as evidence of pragmatism.
  • The stimulus package was a failure, despite being enacted in January 2009 with the prediction “that if the stimulus passed, unemployment would be at 7 percent at the end of 2010.”  Stimulus proponents continue to defend the strategy but say the particulars were simply “inadequate and poorly targeted”.  The inside-the-beltway crowd views this as a problem in expectations-setting rather than actual policy.
  • Unemployment is obviously still distressingly high and government statistics underreport the pain: “Counting those who are seeking full-time jobs while working part time and those how have stopped looking altogether, it’s closer to 17 percent.”
  • The economic team was “fractured” and “the word most commonly used by those involved is ‘dysfunctional’”.  Larry Summers bears the brunt of it (you know him from his appearance in The Social Network).  Budget Director Peter Orszag says, “Unfortunately I think the environment often brought out the worst in people instead of the best in people. And I’d include myself in that.” I have a gift suggestion for the new team.
  • They continue to flail for ideas and a strategy.  As recently as the week before Christmas, the President replies, after being presented with “familiar and uninspired” proposals, with “I’ve told you before, I want you to come with ideas that excite me.”
  • The primary speaking point, voiced by Treasury Secretary Timothy Geithner, is “it could have been so much worse.’”  Despite the litany of screw-ups mentions in the article, no one had the temerity to suggest that it could have been better as well.

My purpose in writing this is not to score partisan points (one can easily argue that Obama’s core economic policy differed little from Bush’s, and both were paltry in impact compared with the Federal Reserve), but rather to indict the whole macroeconomic-industrial complex (a sentence which to me evokes this clip starting at about 1:45).  What set me off is the implication in the article that the administration is looking not just to rejuvenate the economy, but also to salvage the reputation of government management of the economy.

Creating millions of jobs is one thing, but redeeming faith in the Keynesian dream of technocratic micromanagement of something as ridiculously complex as our economy after the last few years certainly qualifies as a big hairy audacious goal.  Especially after the economic policy team responsible admits they weren’t guided by a clear strategy or set of principles, didn’t understand quite what was going on in the economy, implemented a program that was ineffective and badly missed its predicted impact, don’t know what to do next, and were the poster child for a dysfunctional team, there is some real work necessary to believe they’ll get it right next time.  The basic problem is the economic models that underlie all these policy prescriptions seem to work better in every universe except our own.  The Times article mentions a model supporting the “it could have been so much worse” school:

Without the actions taken by Bush, Obama and the Federal Reserve, the economy was headed to what Bernanke called “Depression 2.0,” in which unemployment potentially would peak at 16.5 percent, according to a later study by Blinder and Mark Zandi, chief economist at Moody’s Analytics.

I somehow resisted the urge to pillory this study, entitled “How the Great Recession Was Brought to an End”, when it first came out.  The New York Times’ lede at the time said:

“Like a mantra, officials from both the Bush and Obama administrations have trumpeted how the government’s sweeping interventions to prop up the economy since 2008 helped avert a second Depression.  Now, two leading economists wielding complex quantitative models say that assertion can be empirically proved.”

Empirically proved!  You see, Misters Blinder and Zandi have a model of the US economy.  They can type in some parameters and find out what would have happened in the absence of the fiscal and monetary actions of the last couple years.  This model is so good, it can tell us how much worse it could have been to three significant figures.  They “empirically proved” that by the end of 2010, real GDP is 6.61% higher, there are 8.40 million more jobs, the unemployment rate is 5.46 points lower and the Low Income Home Energy Assistance Program has a Keynesian multiplier of 1.13 thanks to various government actions.

Conveniently, this counterfactual is set in an alternative universe which no one can disprove.  When it comes to forecasting things in our universe, this model and its cohorts don’t do so well.  This model that so accurately predicts events in that alternative universe didn’t predict the global financial crisis nor does it accurately predict what will happen next in our universe.  And there seems to be some kind of agreement amongst polite company not to point out the huge role of failed economic models in causing the global financial crisis (Michael Lewis’ The Big Short is a great read and a very accessible introduction to wayward economic models).  And I probably shouldn’t point out that Zandi and this model both come from Moody’s which of course used its various models to rate all those mortgage securities that blew up as AAA risks (and Moody’s of course has no incentive to pander to the government in order to keep their Federally mandated position in the bond rating oligopoly…).

The epistemic arrogance, to use Nassim Taleb’s phrase, of the macroeconomics profession is staggering.  They still have the keys to the car and are trying to pass their driver’s test by pointing to how they would have parallel parked in another universe, despite having hit the cars in front and behind them as well as sideswiping the parking meter in this universe.

The Times story does offer two bright spots.  One is perhaps they have figured out economic growth is the only hope.  The new head of the Council of Economic Advisors says “We’ve shifted out of the rescue mode.  We’ve got to move into full-fledged growth mode.”  And This Time is Different: Eight Centuries of Financial Folly makes an appearance.  I highly recommend this book, even if you just read the first and last two chapters (though you’ll miss cool things like how Newfoundland lost its sovereignty and the fact Greece has been in default roughly every other year since it gained its independence from the Ottoman Empire).  This Time is Different suggests tendencies in economic behavior and the consequences of various government policies without pretending to be able to accurately predict or control them.  The concluding paragraphs:

“The lesson of history, then, is that even as institutions and policy makers improve, there will always be temptation to stretch the limits. Just as an individual can go bankrupt no matter how rich she starts out, a financial system can collapse under the pressure of greed, politics, and profits no matter how well regulated it seems to be.

Technology has changed, the height of humans has changed, and fashions have changed. Yet the ability of governments and investors to delude themselves, giving rise to periodic bouts of euphoria that usually end in tears, seems to have remained a constant. No careful reader of Friedman and Schwartz will be surprised by this lesson about the ability of governments to mismanage financial markets, a key theme of their analysis.  As for financial markets, Kindleberger wisely titled the first chapter of his classic book “Financial Crisis: A Hardy Perennial.”

We have come full circle to the concept of financial fragility in economies with massive imbalances. All too often, periods of heavy borrowing can take place in a bubble and last for a surprisingly long time. But highly leveraged economies, particularly those in which continual rollover of short-term debt is sustained only by confidence in relatively illiquid underlying assets, seldom survive forever, particularly if leverage continues to grow unchecked. This time may seem different, but all too often a deeper look shows it is not. Encouragingly, history does point to warning signs that policy makers can look at to assess risk – if only they do not become too drunk with their credit bubble-fueled success and say, as their predecessors have for centuries, “This time is different.”

In the meantime, I await the Nixonian proclamation that “We are all Hayekians now.”

Fun with Numbers: State Fiscal Policy Edition

Warning: serious wonkery ahead.  Not for the faint of chart.

Amid discussions of how dire are the fiscal situations of American states and which state most resembles Greece, plus various initiatives on the ballot in the recent election, I was curious how our fair state of Washington stacks up in terms of fiscal responsibility and sustainability.

I pulled spending data on California, Illinois and New York as they’re the most often compared to Greece (Athens-on-the-Pacific/Great Lakes/Hudson respectively).  I also threw in Texas which has outperformed most states economically over the last ten years without being completely dependent on natural resources (sorry Dakotas) and it supposedly has a different governmental philosophy (does “everything is bigger in Texas” extend to fiscal policy?).

The results are surprising as seen in Figure 1 below.  Since 2000, Washington state government spending has grown faster than that of California, Illinois and New York (gulp!).  And that is after being the most frugal through 2006!  New York is actually the most restrained state of the bunch.

Figure 1: Growth in State Government Spending: 2000-2010
(Data is normalized so 2000 spending = 100)

If you look at it just since 2006 in Figure 2, Washington’s spending growth dramatically exceeds the other Hellenic wannabes.  Spending is up almost 60% in five years.  (And note supposedly modest Texas has seen faster growth than California in this period).

Figure 2: Growth in State Government Spending: 2006-2010

My operating assumption is government spending can’t grow much if at all faster than the underlying economy over time.  Unfortunately, that isn’t even remotely the case in Washington, as seen in Figure 3.  Growth in state spending has outpaced the economy massively over the course of the last decade.  The economy has grown by a little over 40% while state government spending has grown by almost three times the economic growth rate.  Now you could argue the growth early in the decade is the result of countercyclical spending and automatic stabilizers in response to the recession after the dot com collapse, but that doesn’t explain more recent growth.  Spending hit an inflection point in 2006 and has grown dramatically faster than the economy since.  This takeoff predates the Second Global Contraction or the Global Financial Crisis or the Big-Ass Recession or whatever we’re supposed to call our current predicament, so it isn’t just countercyclical spending.  In fact it is a straight line up from 2006.

Figure 3: Growth in Washington State Government Spending and Washington State Economy: 2000-2010image_thumb10

For comparison, the same spending vs. economy comparison for California is in Figure 4.  It is a similar graph of excess although California is less of a hockey stick than Washington as they lacked the restraint of Washington in the earlier years.

Figure 4: Growth in California State Government Spending and California State Economy: 2000-2010  image

Now normalizing the data above to show growth hides some interesting comparisons between states.  Washington is the smallest of the states in the sample, with an economy half the size of Illinois and a sixth that of California.  But in relative terms, state government spending consumes a higher percentage of the Washington economy than any of the other four states as seen in Figure 5.  Moreover, it has grown from 10% in 2000 to now almost 14%.

Figure 5: State Government Spending as a Percentage of the Total State Economy: 2009

When you normalize spending for economic growth in Figure 6, allowing faster growing economies to grow their spending faster, California regains the lead for the most growth in state spending relative to underlying economic growth.  Both California and Washington grew state spending relative to economic growth by about three times what New York and Texas did over the last decade.  Washington has had decent economic growth during most of the last decade and outgrew both California and Illinois in absolute terms, but not Texas or New York.  But Washington has been growing spending much faster than any state since 2006 relative to economic growth.  And just to be clear, on this graph, government spending growth exactly in line with economic growth would be a perfectly flat line at 100, so other than Texas briefly, all our states have seen spending significantly outgrowing their economy.  You do see what looks like a countercyclical upswing in spending after the 2000 downturn, but the next upswing starts before the current downturn and just keeps on going even as the economy goes from growth to recession.

Figure 6: State Government Spending Normalized for Economic Growth: 2000-2010image_thumb7

I’m not sure what to conclude other than Washington state spending growth over the last decade seems irresponsible and unsustainable.  Washington was on a relatively modest path but then went nuts in its spending after 2006 and I don’t know why.  Strong economic growth is the absolute best solution for fiscal challenges, but even then you’d like to see growth in government spending tracking the rate of economic growth, and certainly not growing at three times the rate.  Unfortunately, we’re in all likelihood in for a period of slow economic growth over the next few years (read This Time is Different: Eight Centuries of Financial Folly for a good, if depressing, statistical look at the aftermath of financial crises – you can just read the first and last two chapters and skip all the data exposition in between, though you will miss cool asides on Henry VIII’s fiscal policy and how Newfoundland lost its sovereignty.  It also reinforces, starting with the title, the inability of government policymakers and appropriators to learn from past consequences of irresponsible fiscal behavior). 

A Note about Data

The data came from which lets you download raw data for manipulation.  Some of their data is historic, some estimated and some interpolated.  I double checked the data with Federal and individual State sources and, interestingly, none of the numbers matched identically but all were close.  I assume any noise averages out.  To get a full economic cycle, I started with the year 2000.  And these numbers don’t include various off-budget shenanigans or underfunded pension funds or any of the other inexorable headlines of the near future.

My Excel spreadsheet with the underlying data and normalization calculations can be downloaded here for those of you who want to wonk out even further.  And I uploaded it to both Google Docs and Microsoft Office Web Apps so you can have your own personal battle of web productivity solutions.

It would be interesting (but not interesting enough for me to actually do it today…) to look at a couple other things:

  • Compare these spending numbers against other measures of economic performance such as per capita and household income as well as disposable income. 
  • Combined state and local government spending to see if overall spending per head between states is closer. 
  • Where the spending growth came from?  The dataset above lets you break down spending into some standard categories (e.g. healthcare vs. infrastructure).  Did specific categories booming or was it growth across the board?
  • Why did Washington State spending go nuts after 2006?
  • What has happened on the revenue side? How big a gap is there relative to spending as state revenues have declined along with the economy?  This is the biggest factor in sustainability – how well has Washington revenue held up relative to other states?

Hoping for Change: Monetary Policy Edition

Platformonomics does not usually do monetary economics (except maybe the virtual kind), but I just happened to be thinking about the Federal Reserve yesterday (admit it, so were you…) shortly before the news broke of Bernanke’s renomination.  I was hoping for a more radical announcement:

Software Bot To Be Nominated Chairman of 
Federal Reserve System

Cutting edge technology tapped to bring stability and
consistency to monetary policy

WASHINGTON D.C. – August 25, 2009 — President Barack Obama today announced he intends to nominate Monet 3.0, a software bot, to become Chairman and a member of the Board of Governors of the Federal Reserve System. As Chairman, Monet will be charged with conducting the nation’s monetary policy by influencing money and credit conditions in the economy.

“Software bots today are successfully outperforming the world’s best human practitioners in complex endeavors like chess, and they do so without irrationality or exuberance,” said President Obama. “Despite the bot’s French-sounding name, I am confident that Monet 3.0’s discipline and transparency will bring price stability and foster the economic growth required for a full economic recovery.”

Monet was born in a lab at Stanford University in the early 1990s and is currently in version 3.0.  The software instantiates a modified version of the Taylor Rule.  Source code for Monet is available for broad inspection and reuse under the modified BSD license at

Monet’s nomination requires approval by the Senate and a bout of hyperinflation that makes the Hungarian episode of 1946 look modest.

It’s the Balance of Profits, Stupid

More fodder for the irrelevance of trade deficit figures in a piece by Hal Varian in the New York Times via a reader of Greg Mankiw’s blog (whew!) on the contributions of various countries involved in the manufacture of the iPod and the associated accounting in the trade statistics:

I found this article by Hal Varian, and a sentence at the end caught my attention. It says that although the Chinese only add 1% of the iPod’s value, each unit exported to the US contributes about $150 to the bilateral deficit. This left me wondering: could bilateral trade numbers, namely the US deficit with China, be a fiction?

Mankiw’s response focuses on the meaningless of bilateral trade deficits but the broader question is how meaningful are aggregate trade deficits or surpluses?  Just as you wouldn’t judge a company without looking at its profits, you can make the same argument for nations.  China looks remarkably like Japan pre-1989: big surpluses, but tiny or non-existent profits.  Don’t tell Lou Dobbs.

The money quote from Varian:

Those clever folks at Apple figured out how to combine 451 mostly generic parts into a valuable product. They may not make the iPod, but they created it. In the end, that’s what really matters.

The iPod example supports the arguments made by Andy Kessler (whose How We Got Here is the best cutting room floor product ever, assembled from a lengthy aside on the history of technology and markets cut from another of his books) and “Mr. Black Swan” Nassim Nicholas Taleb that I mentioned previously:

He [Taleb] arrives at a similar conclusion as Andy Kessler on globalization and division of labor.  Don’t sweat the US’s trade deficit.  That is just revenue: just look at the balance of profits (where the US runs a surplus).  He believes the US has focused on scaleable businesses where your revenue is not limited by your number of labor hours, but rather those that involve creativity and are often winner-take-all in the global economy.  We export jobs for the non-scaleable elements to others who are happy to be paid by the hour: “There is more money in designing a shoe than actually making it; Nike, Dell and Boeing can get paid for just thinking, organizing and leveraging their know-how and ideas while subcontracted factories in developing countries do the grunt work and engineers in cultured and mathematical states do the noncreative technical grind”.

I also remember a wry comment in The Economist years ago that if you added up all the reported trade balances, the world ran something like a $70 billion trade deficit with itself (cursory search on the site doesn’t turn it up, but they say things like “trade deficit” and “$70 billion” a lot).  So not only is it a questionable metric to act upon, but the data itself may be questionable as well.