Nostradoofus

The Greek financial crisis is a mere deck chair on a political Titanic: European monetary union is inherently unstable without fiscal and central bank union. Fiscal/central bank union would require political union, a political impossibility. Thus, EMU cannot move past its current unstable state. Retreating to a mere trade union offers most of the benefits of EMU, at a fraction of the complexity and distortion.

Thus, almost inevitably, the euro area will shrink or break up, while the EU trade area will thrive.

Why is EMU inherently unstable? Because it is, by its nature, a transitional condition. Fiat currency, central bank and fiscal policy are logically a single module: a governing entity manages its fiat currency through fiscal and monetary policy to achieve domestic political goals (rational or irrational, short or long term oriented).

If a country tries to use someone else’s fiat currency without control of the monetary and fiscal policy of that currency (as occurs, for example, when Argentina dollarizes or when China pegs, or when Greece euro-izes), the result, every time, is systemic imbalance: inflation/deflation, current account deficit/surplus, etc. This happens because the solution to such imbalances is to revalue the currency — but they can’t, because it’s someone else’s currency.

To eliminate these distortions, the inevitable end is always either to unify more (merge politically) or less (stop using someone else’s currency).

Some EU framers understood this, and from the beginning intended EMU to be a transitional step toward political union. However, as it turned out, EMU was opposed by a popular majority in nearly every country where it was adopted. Going further — giving up the national budget to Brussels — is obviously impossible in the foreseeable future. So one has to ask if an inherently transitional state, with its inherent distortions, can persist indefinitely.

At this point, it makes sense to ask why Europe ever wanted this. The answer is rooted in World War II. When European politicians hatched these ideas in the early 1950s, militant nationalism was a vivid memory. The intended benefits of the EU were something like these:

1. Minimize cross-border transaction costs.
2. Present a united geopolitical force vs the USSR.
3. Prevent nationalist warfare among European states.

But in the 60 years it has taken to get this far, conditions have completely changed. There is little external military threat. Intra-European warfare sounds quaint. Europe’s trade union, coupled with modern IT/comms technology, has already greatly reduced cross-border transaction costs.

Under modern geopolitical conditions, a mere trade union may be Pareto-optimal: with a fraction of the complexity and distortion of a currency union, a trade union achieves most of the economic benefit, provides most of the necessary bargaining power vs Russia, yet avoids, for example, the inherent division between fiscal management styles of Mediterranean and non-Mediterranean countries.

A partial euro breakup may thus be inevitable, but, given the impossibility of political union, may also be for the best.  In essence, Europe is doing something today that makes no sense — just because it made sense 50 years ago.

—

A popular finance blogger asked me yesterday where Palm went wrong. Though I was a leading developer of Palm games and graphics software, I hadn’t considered this in a while. The specific question might be:

Why does the iPhone earn a third of all smartphone segment profits on just 8% share (by revenue), while Palm, despite an 11-year lead in mobile devices, is on life support, with negative lifetime profits, its shares down a split-adjusted 99.26% since the IPO?  What happened?

The answer is prosaic:  profit requires a sustainable competitive advantage.  Palm never found one.  Apple did.

One could kvetch all day (all week) about operational errors at Palm, but the central problem was strategic.  They never raised barriers to entry.  The carriers, retailers, distributors and PC vendors always held the power. Palm remained just a tool — though a very good one — for replicating other vendors’ PC data in the field.

Many bought the iPhone because they already own a pile of music purchased through iTunes — music that can only be played through an iPod.  You don’t want to carry a separate phone and iPod, so you pay a premium for an iPhone.  In short, the switching costs and network effects of iTunes are a big reason iPhone is so profitable.

Also when you buy an iPhone, likely as not you buy it directly from Apple.  There is no 3rd-party distributor or retailer to eat up margin.

Yes, iPhone is a great product, but that is not what sustains the margins.  The switching costs and distribution control sustain the margins.

Palm never grasped this. They still don’t.  They are narrowly focused on product. Their original design was fantastic, and opened a whole product category. The Pre is also great, by all indications. But good design is merely a necessary, not a sufficient, condition for consistent profitability.

Lacking any defenses, Palm still must fight every day for retail shelf space and wholesale pricing; Apple need not.  Palm cannot cross-sell from desktop computers; Apple can.  Palm controls no meaningful content format;  Apple does.  There is a long list like this, and it accounts for Palm’s travails as much as, or more than, any operational limitation.

For a time, perhaps 1996 to 2003, Palm had a big tactical advantage.  Their original design was so far out in front that competition was fragmented and poorly orchestrated.  That time could have been spent developing defenses.

For example, Palm could have used the PalmPilot sync software as a beachhead to wrest control of PC contact management software from ACT.  Palm could then have mainted those contacts in a closed data format, creating switching costs.  As long as their software was truly excellent, industry beating, then customers would be happy, despite a premium price.  Instead, Palm abdicated, treated contact management software as an afterthought for 10 years, while third parties developed sync tools.

The tactical advantage is now long lost.  Game over, it seems.

—

Following up on recent posts about inefficient regulation, here are several specific examples to show how California could radically simplify its operations.  This would slash state expenditures, hugely reduce compliance demands on small businesses, reduce barriers to hiring and business formation, yet have little or no impact on existing tax and revenue structure.

1. No state-specific income tax rules.  Conform state tax to federal, without exception.  You simply pay 10 cents to Calif. for every dollar you pay to the IRS, minus any amount you paid to other states.
2. No “use tax.”  Place burden on the seller to charge and remit California sales tax for California buyers.
3. All state and county interactions are online and paperless by default:  taxes, licenses, etc.
4. No smog check on vehicles under 10 years old.  Gross polluters were all made before 1980 anyway.
5. Eliminate state investment advisor registration.  This is how most other states do it, because SEC already registers managers with over $30m AUM.  This is a classic 95/5 Pareto optimization:  by dollars, most fraud is concentrated in big scams like like Madoff and Allen Stanford;  meanwhile, registration has not been shown to reduce fraud, and the law still applies whether registered or not.

—

Pundits endlessly debate who the survivors will be among traditional news sources.

It’s quite possible the answer is none.  No for-profit survivors at all.  An economist should agree this is a realistic possibility for a perfectly competitive industry with marginal costs of zero.

We have forgotten why newspapers first arose:  printing and delivery are expensive.  This meant an aggregator like the New York Times had much lower costs than a lowly pamphleteer.

This has been turned on its head.  The pamphleteer now has the lower costs, because he needn’t employ an office, printing press, trucks and union labor, as the NYT must.

In the future, news may instead be gathered mainly by nonprofits or individuals.  It may be edited by respected, subject-specific independent editors — some sophisticated future variant of today’s best bloggers, e.g. CalculatedRisk.

This particular outcome is not certain — not as certain as death, taxes and the inevitable collapse of the Los Angeles Times print edition.  But it is realistic.

—

When neither political party represents the public interest, you get nonsensical divide-by-zero errors in public sentiment.

For example, it seems the American people are so angry about the biggest wealth disparity in a century that they — suddenly are starting to like Republicans again.

You can’t make this stuff up.

—

From the 16th century until about 40 years ago, “refute” had only one definition:  ”to disprove by reasoned argument or evidence.”

Linguists began to complain in the 1960’s that “refute” was increasingly used as a synonym for “deny.”  Evidence or argument was no longer required.

The latter usage became widespread in the news media in the past decade — so widespread, in fact, that it is now included as a second definition in many dictionaries.

Of course, plenty of errors and malapropisms have found their way into common English over the same period.  The nonsensical malapropism “hone in” (sharpen inward?) is now universally substituted for the original “home in.”  The misspelled “supercede” is now an acceptable substitute for the original “supersede.”  Words ending in “own” are increasingly pronounced as two syllables, i.e. “known” sounds like “knowen,” “shown” sounds like “showen.”

But unlike those harmless cases, the redefinition of “refute,” which conflates reasoned argument with simple denial, can be seen to symbolize the general decline of reasoned argument in the Anglophone world over the past generation.

—

The previous post on modular bureaucracy can be summed thus:  to instantly reduce unemployment, radically simplify small business employment rules to make hiring effortless. Here is a specific suggestion.

For businesses with less than $500k gross receipts and fewer than 3 employees, all of whom are paid less than $80k/yr:

• No federal withholding (employee is responsible for his own taxes).
• Payroll tax, social security and medicare are a flat $300 per month per employee, payable by monthly automatic ACH (supported by all US banks), no paper checks.
• Federal govt provides first $100k of liability and worker’s comp insurance.

This is a classic 80/20 optimization, or more like 95/5.  Small business hiring would explode. I suspect it would bring some of the shadow economy (cash employment off the books) back into the legitimate system.

People do not appreciate the scale of the benefit this change would bring.  A huge percentage of total employment in the US is from the smallest businesses.  This move could prompt millions of sole proprietorships to hire their first employee.

Yes, it would turn worker’s comp into even more of a fraud machine — but that is a problem with worker’s comp law, not with insurance or employment law.  A properly modular bureaucracy would solve that problem separately.

—

The news story du jour is to muse about why small businesses are not hiring.  Talk of tax credits, in my view, totally misses the point.

Nostradoofus essentially answered this question some time ago:  the problem is not cost, but red tape, both government and private.

The non-salary costs of hiring employees (chiefly health insurance, worker’s comp, liability insurance, legal work and tax filings) have grown far more complex and unpredictable in the past 15 years, yet are almost entirely outside the control of both employer and employee.

The problem is not so much the expense, but the trend toward ever greater UNCERTAINTY and COMPLEXITY of employing workers — the thousand small details that constantly change and that are outside the employer’s control.

This problem falls disproportionately on small business, which lacks the scale to employ specialized human resources staff to handle the complexity and unpredictable change.  Small business also lacks the financial depth to handle unpredictable changes in cost.

Repeat:  the costs themselves are not the problem.  Health insurance and pensions, for example, are both good and necessary. The problem is the UNCERTAINTY of those costs, and the COMPLEXITY of compliance.

This trend also helps explain offshoring.

To bring the problem into better focus, consider the fact that most small businesses have 0 or 1 employees.  As a result, to move the unemployment needle significantly would require that we convince a lot of solo businesspeople to hire their first employee.  This, in turn, would require convincing each of those prospective employers to do all of the following.

• File weekly, monthly and quarterly employment tax reports with at least 3 different taxing agencies.
• Expose themselves to huge penalties if they file anything incorrectly.
• Educate themselves about insurance (liability, worker’s comp, medical) and pensions.
• Shop for insurance at least once a year.
• Take on “single unknowns” like unpredictable growth in health insurance costs.
• Take on the “double unknown” of unpredictable liability exposure to employees.

Balanced against this commitment of hundreds of hours a year and a totally unpredictable financial commitment, the prospective employer has a simple alternative:  bid the work out on Elance.com.  If the employer can engage someone outside the US, they eliminate all tax filings, insurance and pensions, and almost all legal liability.  They just send cash by PayPal when the job is done.

From the perspective of the harried, overworked solo businessperson, this simplification is much more compelling than any mere cost advantage from offshoring.  Other things equal (cost and quality), offshoring is by far the better deal for the small business, because it is so much simpler.

The only policy solution here is for the US to get serious about streamlining its sclerotic employment system.  For example, one of the best arguments for nationalized health care and pensions is that they are simple and modular.  You just pay into them and get the services.  This allows both worker and employer to focus their attention on other things.

Independent modules, by their nature, offer lower performance than purpose-optimized solutions.  Viewed in isolation, modules are suboptimal.  But what they lack in efficiency, they more than make up for in simplicity and maintainability.  This is as true with government and bureaucracy as it is with software development.

It would be irrational to argue that optimization is always more important than simplicity, just as it would be irrational to argue we should write all software in assembly language, just because it will run faster.

—

One of the main reasons Clinton was able to balance the budget in the 1990s was that Treasury Secretaries Robert Rubin and Larry Summers restructured U.S. sovereign debt to shorter maturities. Did this unintentionally create the conditions for artificially low interest rates, serial bubbles and sovereign debt spiral?  Read on…

Readers are probably aware that the single most important factor in Clinton’s budget surplus was the Treasury’s move to shorter-dated bonds. Short durations almost always have lower interest rates (except when they don’t — more on that in a moment). Governments have traditionally borrowed almost entirely in long-term bonds, such as 10-year and 30-year Treasuries. In the 1990s, the U.S. moved to shorter durations, resulting in instantly lower debt service costs, and a balanced budget.

The risk, which I’ve not seen mentioned anywhere, is that the federal government then becomes remarkably similar in capital structure to an investment bank, depending upon constant refinancing of huge short-term debts just to remain solvent — exactly the capital structure that killed both Bear Stearns and Lehman Brothers. With the U.S. government, it seemingly plays out in two ways.

1. Massively politicizes the Fed.  The Chairman of the Federal Reserve (who is unelected) now directly controls the size of the U.S. budget deficit simply by changing short-term interest rates, which changes the cost of the next rollover. Or, viewed from the other side, there is unprecedented pressure on the Fed to hold rates artificially low, to manage the deficit. (Maybe the serial bubbles and artificially low rates since 1996 result largely from this new pressure.)  This pressure did not exist before, because long-dated bond rates are determined by market prices, not by the Fed.

2. Massively exposes the US to a sovereign bond panic.  Suppose everyone decided for a couple of years that US T-bonds were toxic (perhaps due to war, terrorist attack, or some other unanticipated event).  If the government were mainly financed with 30-year bonds, then no big deal, they can just wait it out. But if the bonds are largely 1-year maturity, there is no way to roll over the bonds, and financial catastrophe ensues.

This is a particularly huge black swan — the move to short-dated bonds created a short-term illusion of stability, while massively increasing long-term exposure to a sovereign wipeout.

It is also very hard to get out of this situation, because the deficits are too large to roll back into more expensive 30-year bonds — just as a homeowner with an option-ARM mortgage is dreadfully exposed to interest rate swings, but cannot afford to refi back into a 30-year fixed.

The biggest danger seems to be that no one even recognizes the risk. According to Nassim Taleb, this blindness is a key feature of a black swan: unappreciated risk of a rare event with a catastrophic outcome.

Because the situation is apparently inescapable due to fiscal pressure, one might argue the United States is already in a sovereign debt spiral, one that will become visible only when short-term Treasury auctions start to fail.

—

To clarify the net neutrality debate, imagine if Google made a hostile bid for Comcast.

This would place net neutrality opponents in an untenable position.  They essentially say it should be acceptable for a cable company to play favorites in internet service delivery;  but if that’s true, then there is nothing anti-competitive in Google owning a cable company.

Net neutrality opponents can’t have it both ways.  Either biased delivery is anti-competitive, or it isn’t.  If it is, then net neutrality wins. If not, then Google should be allowed to buy Comcast and throttle everyone else’s services.  Which would make no sense.

This is just a specific case of the more general fact that network-effect natural monopolies (railroads, cable companies, phone companies, Microsoft) defy normal microeconomics, because they have increasing returns to scale.  As a result, they tend to require regulation to ensure efficient market function.  In general, this regulation takes the form of forbidding bias in the freight delivered over the network.  So the railroad must carry anyone’s rail car, the phone network must carry any long-distance carrier’s call, Microsoft’s applications division should have been forced to split off from the OS division, and fiber networks should be forced to carry anyone’s traffic.

This applies only to network-effect natural monopolies, and not to other, weaker forms of market power.

—