201710

October 30, 2017 12:37PM

How Time Can Move Backward For Immigrants Waiting for Green Cards

Many Americans want immigrants to “get in line.” But they cannot do so on their own. They need to get a sponsor, either a U.S. citizen family member or a U.S. employer, to petition the government to grant them permanent residency (a “green card”). Even if immigrants do obtain sponsors, there isn’t just one line to get into. Rather, immigrants have separate lines based on the type of sponsorship and their country of origin, and these lines all move at different speeds. Even two immigrants working in essentially the same position whose employer petitions for them on the same day can end up receiving their green cards decades apart if they were born different places.

How America still discriminates based on nationality

This bizarre fact is a consequence of the racist history of U.S. immigration law. In 1921, Congress created the first quota on legal immigration (the “worldwide limit” ). Three years later, it created limits for individual nationalities (the “per-country limits”). The per-country limits give each nationality a share of the worldwide limit. If nationals of a certain country use up their share of the green cards, they have to wait, and immigrants from other countries get to skip ahead of them in line. (And no, Congress made sure that immigrants can’t evade the per-country quotas by getting citizenship somewhere else. Birthplace is all that matters.)

Initially, the per-country limits openly discriminated against “undesirable” immigrants, defined as Asians, Africans, and Eastern Europeans (mostly Jews). But in 1965, Congress made the per-country limits uniform across countries. Today, no country can receive more than 7 percent of the worldwide limit in any green card category. But this reform just shifted the discrimination toward nationalities with the highest demand for green cards. The goal here was not any less racist. The debates over the law abounded with “liberals” reassuring conservatives that America wouldn’t be flooded with Asians.

In order to apply for a green card, green cards must be available under both the worldwide limit and the per-country limit for the relevant category. After their sponsors petition for them to receive a green card, immigrants wait in line to apply for the green card themselves. The State Department releases a green card bulletin every month to inform immigrants of which ones can apply that month. Immigrants whose sponsor petitioned for them before a certain date—called the “priority date”—can apply. Everyone else must continue to wait.

Right now, for example, Filipinos can apply for a green card this month if their U.S. citizen siblings petitioned for them before June 8, 1994—23 years ago. But here’s the critical point: this “priority date” tells Filipinos nothing about how long they will have to wait if their sibling petitioned for them today. If a lot fewer U.S. citizens applied for Filipino siblings after June 8, 1994, they might be able to receive their green cards a decade or more sooner than those receiving them today. However, if the number of petitions increased, then the wait could be even longer—maybe decades longer.

For example, the priority date for Filipino siblings of U.S. citizens in June 1994—when those who are today receiving their green cards started the process—was June 1977. In other words, Filipino siblings filing green card applications in June 1994 had waited from 1977 to 1994. The U.S. citizen who filed an immigrant petition on June 8, 1994 may have looked at the green card bulletin and thought his Filipino sibling would have to wait “only” 17 years. In fact, he had to wait 23.

How green card time moves backwards

To know when an immigrant entering the process today will receive a visa, the government would need to know the number of petitions filed in each year under each green card category, the nationality of the beneficiary of each petition, and the nationalities of any spouses and children of the petition beneficiary.

A precise estimate is impossible because some applicants abandon their petitions, apply sometime after their priority date comes up, have additional children, get married, become ineligible, etc. But one would think that the government would attempt to track the basic information carefully, so it could give at least a decent estimate of the future wait times. But being the government, it naturally doesn’t, so whenever the State Department moves up the “priority date,” it essentially guesses how many people more will apply. If the State Department guesses wrong, it realizes its mistake and moves the priority date back in time. In other words, green card waiting time doesn’t move linearly like real time does. It stops, starts, and even runs in reverse. The government calls movement backward a “retrogression.” 

Figure 1 below highlights how priority dates move in the green card backlog for Indian college graduates sponsored by U.S. employers under the employment-based third preference category. When priority dates move ahead in time, the orange line goes up. When they move back in time, it goes down. When the priority date is the same as the current date for multiple months, it goes in a straight line at about a 45-degree angle, progressing steadily upward with each month. As Figure 1 shows, from October 2002 through December 2004, priority dates were current.

Immigrants who were the beneficiary of a green card in December 2004 would have thought, if they had looked only at the priority date, that they would receive their visa almost immediately. In fact, they had to wait more than a decade—until September 2015 to apply. Then, when they finally did apply, the priority immediately retrogressed again. When this happens, the government sets their applications aside until the date becomes current again. Since December 2004, there have been seven significant retrogressions and several other smaller ones.

Figure 1
Priority Dates for Employment-Based (EB-3) Immigrants from India, October 2002 to November 2017

Media Name: prioritydates.png

Source: U.S. Department of State

The big retrogression in 2005—when the priority date moved back to 1998—isn’t quite as meaningful as it appears. The State Department moved the dates back that far just to prevent anyone from applying. It’s not that the wait really grew quite that much. (If you want to know why the big jump happened, skip to this endnote.[1]) In any case, starting from March 2006—right after the big dip where Figure 1 shows “1-Jan-01”—green card time moved about half as fast as actual time. Priority dates have advanced five years and ten months, while actual time moved forward 11 years in eight months. Figure 2 shows the wait for Indian immigrants to apply more than doubled from 5 years to 11 years, while waits for all other immigrants (excluding China, Philippines, and Mexico) have disappeared.

Figure 2
How Long Applicants Had to Wait to Apply for Green Cards from India and Elsewhere*, October 2002 to November 2017

Media Name: prioritydates2.png

Source: U.S. Department of State
*Excluding China, Philippines, and Mexico


Wait times are actually longer than Figure 2 shows. Figure 2 only shows the wait to apply, not to receive an approval. As I mentioned before, when the State Department moves the dates forward, and more applicants apply than there are slots available, their applications are held in abeyance until numbers are available again. This post-application waiting period still happens today. In fact, nearly 140,000 immigrants are waiting at this stage in the employment-based categories, and more than a third of them are Indians. It will take several years to clear just these cases from the backlog.

So how long will Indian immigrant workers have to wait going forward? Again, no one really knows for certain. In 2012, Stuart Anderson of the National Foundation for American Policy estimated 70 years. In 2014, the government estimated that 234,000 high skilled workers were waiting to apply for green cards. The family of the workers, however, use up a little more than half of the green card quotas, so there are probably now roughly half a million green card holders in line.

For the two employment-based categories with the longest waits, Indians can only receive 5,600 green cards (out of 80,000). We would need to know what share of those in line are Indian and what share are waiting in which of the employment-based categories to provide a good estimate of the wait for immigrants applying today. This information isn’t available. But if even half of these workers are Indians in the EB-3 category—which seems very likely given how much longer Indians in this category have already been waiting than other nationalities—then their wait would be about a century.

In other words, it’s possible that almost all of the Indian workers applying today will die before they receive permanent residency, while other immigrant workers will receive their green cards almost right away. This is the system that Congress refuses to reform.


[1] Two factors apparently combined to slam the brakes on Indian workers—and everyone else in the EB-3 green card category—in 2005. First, Congress passed a law at the start of FY 2001 that eliminated the backlog temporarily in 2003 and 2004. The law waived the per-country limits in situations where the worldwide limit for the category otherwise would not be filled, and it temporarily increased the number of green cards by recapturing visas that went unused in 1999 and 2000 (unused visas are a crazy topic for another post).

At the very same time, Congress created a new agency—U.S. Citizenship and Immigration Services (USCIS)—to adjudicate green card applications from temporary workers in the United States, and it was developing new procedures to adjudicate applications. Even though the State Department kept telling immigrants that they could apply, USCIS wasn’t processing them quickly enough. This led to a growing backlog in green card applications. Once USCIS instituted measures to catch up, the State Department realized that it let far too many applicants apply and slammed on the brakes. These applications are then held in abeyance.
October 30, 2017 11:36AM

FCC Finally Deregulates Telegraph Service, 11 Years After Its Demise

“The FCC said in a notice it was removing ‘outmoded regulations’ on telegraphs effective in November.” And none too soon: “The last Western Union telegram in the United States was sent in 2006” and the “last major telegram service worldwide ended in India in 2013.” Reuters reports:

AT&T Inc, originally known as the American Telephone and Telegraph Company, in 2013 lamented the FCC’s failure to formally stop enforcing some telegraph rules.

“Regulations have a tendency to persist long after they outlived any usefulness and it takes real focus and effort to ultimately remove them from the books even when everyone agrees that it is the common sense thing to do,” the company said.

In 2011, I observed that Connecticut had yet to get around to repealing old state laws like those regulating the working conditions of telegraph messengers (cross-posted and adapted from Overlawyered).

October 30, 2017 4:03AM

Supreme Court Should Protect Federalism against Pests Both Rodential and Congressional

Does the federal government enjoy plenary power to regulate every aspect of corporeal existence, down to the rodents living in your backyard? People for the Ethical Treatment of Property Owners (PETPO), an organization of concerned citizens from Utah, say no, and want the Supreme Court to hear them out.

Article I of the Constitution lists the federal legislative powers: Congress may only act pursuant to one of these enumerated powers. One of these powers is the regulation of commerce “among the several states.” Starting with the New Deal, however, Congress has increasingly looked upon that power as a license to do whatever it likes. And for decades, the courts rubber-stamped these increasingly expansive federal intrusions into areas traditionally reserved to the states.

But in a series of cases, starting with 1995’s United States v. Lopez, the Supreme Court began to push back, reaffirming that federal regulation under the Commerce Clause must be, well, commercial. Recall that while Chief Justice John Roberts ultimately saved Obamacare by transmogrifying the individual mandate into a tax, he and the Court majority rejected the government’s arguments regarding the Commerce and Necessary and Proper Clauses.

That brings us to the current case. The Utah prairie dog, which resides only within a small corner of southwest Utah, has no commercial value: there is no market for it—they make terrible pets—or any product made from it. Moreover, the current population is large and expanding. Yet it is listed as “threatened” under the federal Endangered Species Act.

Its legal status derives from the distribution of its population: the government deems the 70% residing on private land a nullity, counting only the federal-land population, on the theory that the citizens of Utah would declare open-hunting on privately domiciled prairie dogs if the species were delisted. And, according to the U.S. Court of Appeals for the 10th Circuit, it doesn’t matter that the varmint is commercially worthless; other unrelated animals have commercial value, so the federal government can stick its nose into whichever animal it likes. Under this theory, of course, all organic life in the United States is subject to congressional whim, because some conjectural private party might impose some vaguely defined harm at some hypothetical future date.

PETPO has filed a petition asking the Supreme Court to review the case. Cato, joined by the Reason Foundation and the Individual Rights Foundation, has filed an amicus brief urging the Court to review PETPO v. U.S. Fish & Wildlife Service. At stake is not simply the beleaguered citizenry of Utah who wish to live their lives unmolested by pests—neither those living underground nor in the District of Columbia—but also the very system of enumerated powers that has protected the liberty of all Americans since the Founding.

October 27, 2017 11:07AM

Consumers Losing Interstate Wine Options

Eric Asimov at the New York Times has an excellent, detailed, and highly discouraging look at the reversal of one of the favorable trends for freedom of commerce in recent years, the greater ease of interstate wine shipment.  Excerpt: 

In the last year or so, carriers like United Parcel Service and FedEx have told retailers that they will no longer accept out-of-state shipments of alcoholic beverages unless they are bound for one of 14 states (along with Washington, D.C.) that explicitly permit such interstate commerce....

Strictly speaking, it was probably never entirely legal in New York or in many other states to have wine shipped in from out-of-state retailers. Yet, these laws requiring a license for interstate wine shipments seemed vague and were rarely enforced....

But now, states — urged on by wine and spirits wholesalers who oppose any sort of interstate alcohol commerce that bypasses them — have stepped up enforcement efforts. Retailers say that the carriers began sending out letters to them a year ago saying they would no longer handle their shipments.

Asimov notes that the cost of restraints on commerce falls most heavily on those who live far from high-end markets: 

For consumers who live in states stocked with fine-wine retailers, like New York, the restrictions are an inconvenience. For consumers in states with few retail options, they are disastrous. It’s hard enough outside major metropolitan areas to find wines from small producers. The crackdown makes it that much harder.

Unfortunately, alcohol wholesalers are among the most powerful of state-level lobbies, and intent on keeping a system that serves their interests. They invoke far-fetched health and safety rationales, claiming that restraints on interstate shipment are "all that protects the wine and spirits business from descending into chaos. The Supreme Court did not buy the argument in 2005, and to me, their economic interest seems a far more likely motivation than public health," writes Asimov.

The Supreme Court's constitutional pronouncements in this area, alas, provide only spotty and indirect protection for consumers' rights to do business with willing providers, concentrating instead on improper protectionism directed by states against other states. Ilya Shapiro analyzed the case law in this 2011 post and related podcast. Brandon Arnold, then Cato's director of government affairs, wrote about a 2010 attempt by wholesalers to get their way through federal legislation. And while state-by-state liberalization efforts are underway in many state capitals, they are routinely stymied by the well-entrenched wholesaler lobby. For more background reading, the California-centric Wine Institute has this FAQ.

October 26, 2017 12:15PM

Blockchain + Gold

The Bitcoin system has the great virtue of securely sending value directly from stranger to stranger. It is open to anyone, anywhere in the world. The sender does not need to trust the recipient, nor any bank or other institution, to accurately record the transfer. The Bitcoin “blockchain” provides a readily consulted online public ledger with immutable records. Transfers are indelibly captured, like flies in amber, and made tamperproof by massive duplication and reconciliation of the ledger over thousands of nodes.

Bitcoin also has well-known limitations as a currency, however. First, it doesn’t scale well. The Bitcoin blockchain can process about four transactions per second, whereas Paypal does hundreds, Visa or Mastercard thousands. The blockchain has become congested as the number of transactions has grown. (Reducing the congestion was the motivation for the proposals to enlarge the block size that recently roiled the bitcoin world.) Validation takes at least ten minutes, longer for more secure validation, and even longer when the system is congested.

Cryptocurrency pioneer Nick Szabo has clearly explained that this tradeoff — high security at the cost of slow transaction speed and low capacity for transaction validations per second — is built into Bitcoin’s massive-duplication design:

Bitcoin's automated integrity comes at high costs in its performance and resource usage. Nobody has discovered any way to greatly increase the computational scalability of the Bitcoin blockchain, for example its transaction throughput, and demonstrated that this improvement does not compromise Bitcoin’s security. … Compared to existing financial IT, Satoshi [Bitcoin’s pseudonymous designer] made radical tradeoffs in favor of security and against performance.

Thus a blockchain system like Bitcoin is not itself capable of quickly processing large numbers of retail payments.

Second, there is the network property of a monetary standard: each of us prefers to be paid in the currency accepted by the largest number of our potential trading partners. This property favors the incumbent standard (the fiat dollar in the US) over both bitcoin and gold. It reinforces the volatility drawback: when your rent and utility bills are denominated in dollars, it is risky to hold a bitcoin balance for the purpose of paying them.

Third, bitcoin’s purchasing power is for now highly volatile.  Broader holding of bitcoin as a medium of exchange would reduce volatility, but all three problems impede that.

These drawbacks have inspired initiatives to combine the benefits of blockchain technology with the use of gold-denominated tokens in place of bitcoin. Gold as a potential medium of exchange arguably has lesser limitations than bitcoin on all three scores. First, its payment processes scale well. Second, its value (in dollars or in purchasing power) is somewhat less volatile over daily to monthly horizons, and is much less volatile over longer horizons. Third, its popularity as an asset in private hands is greater. As of 15 October, 2017, total bitcoin balances are worth $92 billion, whereas worldwide private investment holdings of gold coins, bullion, and ETFs are estimated at $1.7 trillion. Gold holdings are more than 18 times larger; bitcoin holdings are less than 6% of gold in private hands.

Initial popularity matters for gaining widespread use in the face of an incumbent currency. Popular dollarization in Latin America and elsewhere gives us a model of how a non-incumbent currency gains a toehold and then spreads. Popular dollarization typically begins with the dollarization of savings, when the local peso becomes a less reliable long-term store of value than the dollar. Dollarization of pricing and payments spread when the peso inflation rate rises to double digits, requiring more frequent revision of peso prices, and imposing a high cost of holding pesos even from paycheck to paycheck.

Several about-to-launch new projects, described below, hope to create gold-based payment systems, while using some form of blockchain technology to enhance the security of holdings and transfers. If gold-backed accounts or digital gold currency tokens with cryptographically secured transfers are successfully launched, users will be able to adopt a modern gold standard as easily as they can now adopt the bitcoin standard.

Gold holding is already widespread as an investment (a saving and tail-risk-hedging) vehicle, as noted, but convenient gold payment mechanisms have been lacking. The enterprise called E-gold was a prototype — a service for individuals to buy, hold, and easily transfer gold account balances — until it was shut down by US authorities in 2005 for nonconformity with US Treasury “anti-money-laundering” and “know your customer” rules. The upcoming new enterprises all promise to comply with AML and KYC requirements for money service businesses.

A gold-denominated digital payment system will have to operate very differently from bitcoin (or any other cryptocurrency). For this reason it is highly misleading to call it “Cryptocurrency backed by gold” as one promotional article has.

A gold-backed account or digital token rests on a commitment to redemption at par, or a price commitment, by contrast to bitcoin’s commitment to the quantity in circulation. The payment processing system will also be different. It cannot be purely peer-to-peer because it requires a gold vault-keeper or equivalent trusted intermediary to maintain the price commitment. But the use of a single trusted limited-access ledger, rather than Bitcoin’s distributed trustless open ledger with its massive duplication in record-keeping, brings a large advantage in the speed and cheapness of payment processing.

Investment Platforms

The first three projects I will describe do not aim at providing a payment system so much as a low-cost platform for investing in gold. Think of them as would-be competitors to gold ETFs or to bullion warehousing services with easy conversions from and into US dollars, like GoldMoney. I will then turn to projects with more potential to generate a sizable payment system.

1. Royal Mint Gold

On its website, Royal Mint Gold (RMG®) calls itself “The New Digital Gold Standard.” This is misleading because, unlike a gold standard in the usual sense, it isn’t a payment system. It elsewhere more accurately calls itself “an investment product” and “a new, cost-effective, convenient and secure way to trade physical gold” with online access and distributed-ledger transparency.

RMG, which promises to come online before the end of 2017, offers much to interest gold investors. It partners The Royal Mint (hereafter TRM), owned by the UK government, with the Chicago Mercantile Exchange (CME). Both are venerable and credible. (There was talk about privatizing TRM in 2011, but it didn’t happen. As a state-owned enterprise, does TRM enjoy sovereign immunity against lawsuits? I don’t know.) TRM will manage the gold vault, and the CME Group will provide the trading platform for electronic warehouse claims to allocated gold in the vault. The RMG system promises that “For every RMG that’s on the network, there’s one gram of gold that’s sitting in our vault.” A proprietary blockchain will be used to record and track whose gold is in the Royal Mint vault. There will be “live, transparent pricing” on the CME trading platform.

To attract gold investors, RMG promises zero “ongoing” management and storage costs. The RMG webpage compares how value grows with the price of gold under its 0% fees, “giving an investment in RMG a projected higher return than physical gold” held in “a traditional gold ETF [that] is assumed to charge an average 0.4% annual storage and management fee.” But where, you might cynically wonder, do the revenues come from to cover TRM’s vault costs?

Here: You buy in at a premium over the spot price of gold (how high is not yet revealed), so TRM gets some float. (They promise to maintain the premium by buying back unwanted RMG as necessary.) You sell out for a transaction fee. You can cash out, and take physical delivery of gold, only in the form of “physical gold bars and coins produced by The Royal Mint,” for which there is a “fabrication and delivery” fee that is presumably large (not yet specified, and it is not clear whether it will be contractually fixed in advance.)

A system that runs on transaction fees even for internal transfers among account-holders discourages using its accounts as checking accounts.

How big does RMG hope to be? “The initial amount of RMG at launch could be up to $1 billion worth of gold. It will be offered through investment providers. Further RMG will then be issued based on market demand.” For perspective, gold ETFs added $2.3 billion on net in the second quarter of 2017.

To summarize, RMG is not a payment system or a currency, much less a cryptocurrency. It is a warehouse claim to gold in a specific vault. It will be salable to the extent that there are many bidders for claims to gold in that vault, but you can’t transfer it to another RMG holder at a zero transaction fee, like writing a check, the way you could with E-gold.

Incidentally, there are two London wholesale payment systems marrying gold to blockchain. A Bloomberg Markets article explains the business case:

About $27 billion of gold changes hands every day in over-the-counter markets where settlements can sometimes take days, leaving price risk for buyers and sellers. Using blockchain promises more transparency, security and speedier deals.

One project to provide this service is called Tradewind, supposed to launch in early 2018. Another is Bankchain Precious Metals. Tradewind promises to provide “a distributed ledger that will handle trade settlement, account management and record-keeping.” Bankchain Precious Metals promises “the instantaneous transfer of payments and ownership of the bullion stored in various vaults in London.”

2. OneGram

Launched in Dubai, OneGram offers a gold-backed (and Sharia-compliant) cryptoasset with blockchain features. Investors in OneGram will be shareholders in a vault full of gold, and will profit as and when the vault accumulates more gold per share.

Its white paper explains:

OneGram aims [at] using blockchain technology to create a new kind of cryptocurrency, where each coin is backed by one gram of gold at launch. In addition, each transaction of OneGram Coin (OGC) generates a small transaction fee which is reinvested in more gold (net of admin costs), thus increasing the amount of gold that backs each OneGram. Therefore, each OGC increases in real value over time, making OneGram unique among cryptocurrencies.

The vault will be located in the Dubai Airport Free Zone. OneGram promises that it will be audited by PricewaterhouseCoopers.

OneGram promotes its cryptoasset as a payment medium, declaring that the “payment institution license is already in place.” But transfers of OGC will be subject to a transaction fee of 1%. The promoters call the fee “small,” but it is high enough compared to ordinary deposit transfer to discourage using OGC as a payment medium.

How the price of gold will be continuously transmitted to the OGC cryptocoin is unclear, because it isn’t clear how the cryptocoin can be converted into the quantity of gold that it is supposed to represent. OneGram promises to have a “payment gateway” for OGC in Dubai and Abu Dhabi, “with fiat conversion.” But how conversion to fiat will be priced is not specified.

OneGram’s ICO (initial coin offering) ran from May 21 through September 4, 2017. It offered 12,400,786 coins, priced at the spot price of one gram of gold (which averaged around $41 during the period) plus 10%, for total revenue of about $550 million if all the coins sold. A  September 6 press release, which announced that the initial coin sale had ended, curiously omitted mention of the quantity of coins actually sold. It also announced that launch of the cryptocoin has been pushed back from October 2017 to “the first quarter of 2018 to ensure that we launch a solid and secure technology solution.”

Critically limiting the potential of OGC to become an important medium of exchange is the feature that no more OGC will be created even if new adopters want in: “100% of total coin supply is pre-mined.” This means that the size of OneGram payment community in value terms will at most grow only slowly with transaction fees, assuming that the price of OGC remains tied to the value of the gold in the vault.

3. OzCoinGold

Much like OneGram, the Australian/American project OzCoinGold promises a limited issue (in this case 100,000 troy ounces maximum, giving a potential market cap of only $128 million at the recent gold price of $1280 per ounce). As with OneGram, the quantity limit prevents widespread use as a medium of exchange. Each cryptoasset token, labelled OzGLD, will be “100% backed” by gold, but with two catches: only one-third of the gold reserves will be above ground as bullion (the other two-thirds will be the proven reserves of a gold mining company), and the tokens can be redeemed for gold only after five years. Audit reports will be uploaded to a blockchain. Accordingly the main sales pitch is as an investment vehicle: it hopes to be “the easiest, most effective and cheapest way to own or invest in gold.”

Payments Platforms

Moving on to gold-blockchain combinations better designed to be payment services, I consider four, beginning with those farthest from launch.

1. Digix Gold Tokens

Although its software is not yet fully coded, the developers of Digix gold tokens have at least spelled out their concept in detail. Digix is headquartered in London. As explained in a press release on Medium, Digix aims at “tokenizing valuable real-world assets” on the Ethereum blockchain. It “intends to be the first to launch a fully trackable and auditable crypto gold token.” A DGX 1 token “contains the right to 1 gram of gold that is stored in an audited vault.” As a claim to gold, like a transferable warehouse receipt, the token “can be easily traded or pledged against a loan without moving the physical gold” from its vault. Validity of ownership is certified through a “Proof of Asset protocol.”

What exactly does it mean to “tokenize” gold? Consider a universal open shared ledger, running on top of the Ethereum blockchain, that records ownership, and transfers of ownership, of a numbered 10g gold bar stored in a known vault. The gold bar has been “tokenized.” As with a unit of bitcoin, once I record a transfer of ownership to you on the blockchain, you can now further pass on the token, or redeem it, and I no longer can.

In other words, DGX is a spendable digital warehouse claim for gold, with ownership validation on the Ethereum blockchain. Of course, payment by transferring claims to vault gold without moving the gold is pretty old hat. Italian banks were doing it around 1200 AD. What’s new is that these claims are warehouse rather than debt claims, and transfers take place in currency-like fashion on the blockchain rather than by use of named account balances on the books of the depository.

The Digix sales pitch is both to gold investors, and to transactors who want to hold purchasing power in spendable cryptoasset form at least temporarily. Unlike unbacked IOU-nothing cryptocurrencies, DGX tokens are claims to physical gold expected to trade at a price tied to the price of physical gold. Gold exhibits less purchasing power volatility than BTC or ETH.

But if reduced volatility of purchasing power is what you want, why not hold the cryptoasset Tether, the price of which has been held fairly steady (so far) at $1? Some people don’t trust Tether. Tether claims to have 100% dollar reserves parked in audited accounts in licensed banks, but it lacks full transparency and there has been controversy over its terms of service. Digix promises greater transparency: warehousing of gold in vaults that are certified members of LMBA, the London Metal Bullion Association, with “Realtime Transparency; immutable on-chain auditing records for your viewing from Inspectorate and PWC; accessible at anytime, anywhere.”

To warehouse your gold, whether purely for storage or (combined with a transfer mechanism) for use as a payment medium, requires you to pay a fee to cover the cost of storage. A typical arrangement is for the warehouse to deduct a percentage storage fee periodically from each account. Gold ETFs typically charge around 0.4% per year. If there is a transfer mechanism, the warehouse may also charge a transaction fee when fulfilling a transfer request. The planned Digix fees appear to be similar to ETF fees. Storage fees will be 0.4% per year to the vault owner. In addition, an Administration fee of 0.2% per year will be charged by the Digix organization, making total annual fees 0.6%. Transaction fees will be 0.1% of transacted amount.

The medieval Italian banks already mentioned introduced the option of accounts with lower fees for customers who wanted not pure storage but rather transaction services, a way to pay people without lugging gold coins around. Such customers brought in loose rather than bagged coins, and consented to fractional reserves, allowing the bank to cover its (reduced) storage costs by interest earned in lending out most of the gold. The advantage for the bank was of course the interest income on the coins lent out. The advantage to the customer was lower storage and transaction fees. Competition among fractional-reserve banks soon reduced storage fees to zero, and even led banks to pay interest on transaction accounts.

Digix promises to tie the price of DGX to the price of gold the old-fashioned way, by redeemability: “Physical Gold Redeemable at any time at our partnering custodial vaults.” The holder of DGX can “Redeem 100 DGX tokens for 100g of physical gold” in person or by mail. However it has not yet specified whether redemption will be at a zero or a positive price. In traditional banking the redemption fee is zero, but it can be zero in this warehousing system only if storage fees are high enough.

The most impressive evidence that Digix intends to promote DGX as a widely used medium of exchange is that they have partnered with a payment card provider (Monolith Studio, whose platform is called TokenCard) to provide an “Ethereum powered” gold-backed debit card. The announced aim is “to ensure gold tokens can be spent efficiently at minimal cost.”

The intriguing vision of Digix and TokenCard is that people will put themselves on a new digital gold standard. A news account quotes Monolith’s co-founder as saying: “Together with Digix, we will be able to offer one of the only true commodity backed debit cards, and bring back the gold standard in a meaningful way.”

2. Glint

The Glint webpage describes its project as a “new global currency, account and app.” Although it says that the project is "Launching in Q4 2017," no specific roll-out date is offered. Headquartered in London, Glint Pay Services Ltd claims “permission to issue electronic money and provide payment services” from the Financial Conduct Authority under the Bank of England. Its co-founder is CEO of GoldMadeSimple.com, an online bullion dealer.

The project has curiously little press coverage online, only a single article which reads like a paid press-release placement. It is very sketchy on details. “Glint is a stealthy London fintech startup that promises to turn gold into a ‘new global currency’. … Glint will offer a frictionless way to both store and spend your money in gold, including at the point of sale, just like a regular local currency. The bigger picture is that gold historically has been a better storage of value than any government-created currency, and therefore — with the aid of technology — is (arguably) a good candidate for an alternative global currency.”

Obviously more details are needed.

3. DinarCoin (DNC)

Despite “dinar” being the Arabic name for a gold coin (derived from the Roman denarius), DinarCoin (DNC) is not linked to Islamic finance. The parent firm DinarDirham is registered in Hong Kong, with offices in Singapore and Kuala Lumpur. It describes the DNC as a “unique digital currency created … on the Ethereum blockchain. The value of each DNC is based on the worldwide gold spot price. DNC can be used for trading, investment and also to make payments.”

Here’s the sales pitch:

If you’re a person that’s interested in precious metals, but are concerned with storage, security, and actually being able to use your metals as cash for purchases, then DinarDirham is for you. You can actually store, secure, and use your gold on the blockchain, and have the ease and convenience of not needing to have it on your person, and accessing it worldwide in minutes.

It will be

a simpler way to transfer gold, as akin to PayPal with dollars. The aim of the DinarDirham is not only to provide additional value and stability to the coin but also to perpetuate the use of bullion as an accepted form of digital currency.

Of course, physical gold isn’t stored on the blockchain. But the record of a contractual claim to gold can be. The promoters promise that “For the lifetime of a DNC a corresponding value of XAU [physical gold] will be held in escrow.  … [W]hen DNC is created it is registered on the Ethereum Blockchain and the Bitcoin Blockchain. The total amount of DNC in circulation can be verified on either Blockchain, and audited against the total XAU held in escrow … by DinarDirham.” Of course, the accuracy of such a comparison is only as great as the accuracy of the reports of the “total XAU held in escrow.” Unlike consensus-validated bitcoin ledger changes, the accuracy of unilaterally altered ledger entries relating to external facts, like the volume of vaulted gold held by DinarDirham, is not ensured by the blockchain.

Unlike OneGram, the volume of DNC payments has the capacity to grow should it catch on as a medium of exchange. If demand growth begins to push the bid price of DNC slightly above the spot price of gold, either the parent firm or one of its “liquidity providers” stands to make an arbitrage profit by buying physical gold, putting it in an escrow vault, and selling additional DNC into the market, until the premium subsides. The vaults are associated with Associated Bullion Exchange, an electronic exchange for allocated precious metals in storage.

The “redemption” mechanism is not straightforward. DNC “can be redeemed for physical gold,” the website says, via a DinarDirham blockchain-recorded digital asset called a Gold Smart Contract (GSC). Unless 1 DNC can always procure 1 GSC, however, “used to purchase” would be more accurate than “redeemed for.” Another account does say that a DNC holder can “purchase” a GSC and then use it “to collect gold from one of many available vaults.” If in fact 1 DNC trades at a variable price for 1 GSC, which is redeemable for 1 gram of gold, it isn’t clear how the price of DNC is supposed to be pegged to the price of gold.

The announced payment-system plans are ambitious. The CEO says:

We are building an entire ecosystem around DinarCoin — exchange, DinarCoin ATM, merchant gateway and debit cards to allow our users to use their digital assets anywhere in the world. Also, physical Dinar Gold Coin (4.25 grams) is already available to order in South East Asia.

How far along is the project right now? Unclear. It hasn’t posted much lately. DNC isn’t listed on CoinMarketCap. I could not find any report on the value of DNC coins currently in circulation.

4. GoldMint

GoldMint is based in Russia. Surf to its webpage from a US location, and you are immediately confronted by a black drop-down box that declares that you may not invest in its ICO if you are domiciled in the US, Canada, China, or Singapore. Its first phase is an ICO ending this month (hoped-for sales, $49 million) for a token called MNTP. A “prelaunch” coin running on the Ethereum blockchain, MNTP will migrate in Q2 2018 to become MNT — the “stake” in the proof-of-stake GoldMint blockchain — which will process transactions in a gold-backed cryptoasset confusingly named GOLD (all caps).

What is most novel and remarkable about GoldMint is that the parent firm claims to be developing hardware called “Custody Bot automated storage facilities,” which it plans to deploy at pawn shops and shopping centers worldwide. Custody Bots will be “programmed to automatically identify and store gold jewelry, small ingots (up to 100 grams) and coins, without human intervention,” taking escrow custody of them for people who want to take out loans collateralized by the gold. (Loans collateralized by gold jewelry are already popular in India, by the way.) Through the spread of Custody Bot automated storage facilities, according to the GoldMint white paper, the firm hopes eventually to handle the storage of gold reserves worth, in US dollars, tens of billions.

More importantly for currency purposes, Bot-stored gold can alternatively be tokenized and traded as the cryptoasset GOLD, which “will become the trading unit for these operations.” GoldMint promises that units of GOLD will always be “100% backed by physical gold and ETF” that the firm holds.

Because the price of gold is less volatile than the price of bitcoin, the firm’s pitch goes, “Crypto traders and enthusiasts can hedge the risks of storing their assets in [a] highly volatile crypto market environment by transferring their savings to cryptoassets GOLD. [Also:] Low volatile GOLD cryptoassets can be used as a payment unit both for companies and individuals.”

GoldMint will not redeem GOLD for gold at par, but it promises to sell you 1 GOLD cryptoasset for a 5% premium over the London spot price of one ounce of gold, and to buy it back at a 3% premium. Thus the total fee for making the round-trip fiat-GOLD-fiat will be 2%. In addition it will assess an “On-Chain transaction fee” of 0.3%, three-fourths of which will go to the miners on the GoldMint blockchain. These seem like fairly competitive fees.

The most important questions about the potential of GoldMint as an important gold-backed currency are about the trustworthiness of its buyback promise, and the reliability of its blockchain for payment validation.

Conclusion

I am not endorsing or recommending investment in any of these projects. Caveat emptor. But I think the last three listed warrant our attention as attempts, in the spirit of E-gold, to provide modern gold-based payment systems with online access. All three explicitly promise not to hold fractional reserves, and say that you can track the volume of cryptoasset on their ledger to see that it matches the number of gold grams or ounces held in their vaults. But if one of them becomes popular as a one-hundred-percent-reserved  gold payment system, perhaps a subsequent innovator will offer zero storage fees and interest on account balances by re-introducing gold-denominated fractional reserve banking. Such a bank, supposing that it surmounts legal obstacles but lacks government deposit insurance, would have to provide as much transparency as potential clients demand to show that it has enough gold and other liquid assets available to redeem promptly all claims that are likely to be presented.

Stay tuned.

[Cross-posted from Alt-M.org]

October 26, 2017 12:08PM

Labor’s Share of GDP: Wrong Answers to a Wrong Question

A recent paper by David Autor of MIT, Lawrence Katz of Harvard and others, “The Fall of the Labor Share and the Rise of Superstar Firms,” begins by posing a mystery: “The fall of labor’s share of GDP in the United States and many other countries in recent decades is well documented but its causes remain uncertain.”  They construct a model to blame it on U.S. businesses that are too successful with consumers.  

Five broad industries, they found, became more dominated by fewer firms between 1982 and 2012: retailing, finance, wholesaling, manufacturing and services. But those aren’t industries at all, much less relevant markets: they’re gigantic, diverse sectors. Is all manufacturing becoming monopolized? Really? Census data ignores imports, but why ruin this bad story with good facts.

Noah Smith at Bloomberg ran an audacious headline about this tenuous paper: “Monopolies drive down labor’s share of GDP.” Smith writes that, “The division of the economy into labor and capital is one place where Karl Marx has left an enduring legacy on the economics profession.” He goes on to claim that “at least since 2000 -- and possibly since the 1970s -- capital has been taking steadily more of the pie.” Yet, Jason Furman and Peter Orszag found “the decline in the labor share of income is not due to an increase in the share of income going to productive capital—which has largely been stable—but instead is due to the increased share of income going to housing capital.” Depreciation and government, they noted, also gained an increased share (i.e., grew faster than labor income.) 

President Obama’s Council of Economic Advisers, under Jason Furman, nonetheless worried that the 50 [!] largest firms in just 10 “industries” (if you can imagine retailing and real estate to be industries) had a larger share of sales in 2012 than in 1997 (using Census data that excludes imports). They concluded that, “many industries may be becoming more concentrated.” Noah Smith, Paul Krugman and many others have suggested that this nebulous “concentration” allowed monopoly profits to rise at the expense of the working class, supposedly explaining labor’s falling share of GDP during the high-tech boom. A quixotic search for even one actual example of monopoly soon morphed into advice about using unconstrained antitrust to constrain Amazon, which is apparently feared to have monopoly profits invisible to the rest of us.  

Research that starts with such a meaningless question as “labor’s share of GDP” was never likely to lead us to any profound answers. Workers do not receive shares of GDP – they receive shares of personal or household income.   

Contrary to popular confusion, dividing employee compensation (wages and benefits) by GDP does not measure how a capitalist private economy (e.g., “superstar firms”) divides income between labor and capital. Most obviously, the government makes up a huge share of GDP, including nonmarket goods like defense and public schools. Nonprofits also account for a lot of GDP, with no obvious payout to labor or capital. Less obviously, depreciation makes up another huge share of GDP, including wear and tear on public highways and bridges as well as private equipment, homes, and buildings. The “imputed rent on owner-occupied homes” is another large piece of GDP. Asking if labor is getting a fair share of defense, depreciation and imputed rent is a truly foolish question. Net private factor income would be a better gauge than GDP, for the purpose at hand, but still flawed.  

The ratio of compensation to GDP uses the wrong numerator as well as an untenable denominator. Labor income must add the labor of self-employed proprietors.

When people say “labor’s share is falling,” they surely mean income people receive from work has not kept up with income people (often the same people) receive from property: dividends, interest, and rent. But, that crude Piketty-Marx labor/capital dichotomy ignores another increasingly important source of personal income: namely, government transfer payments from taxpayers to those entitled to cash and in-kind benefits.  

Shares of Personal Income

The first graph shows shares of income from labor, property, and transfers. The property share peaked at 21.1% in 1984-85, as the Fed kept interest rates very high, but averaged 19.3% and was 19.4% in 2016 (after dropping to 17.8% in 2009).  The labor share averaged 66.5% but was 63.3% in 2016 even though property owners’ share was virtually flat. What went up? Transfer payments. Transfers rose from 11.7% of personal income in 1988 to 17.4% in 2016. Personal income that has been growing persistently faster than income from work has not been income from property (since the 1980s), but income from Social Security, Disability, Medicare, Medicaid, EITC, TANF, SNAP, SSI, UI, and so on.

Some might object that personal income leaves out retained corporate profits. But profits not paid out as dividends add to people’s income only if they are reinvested wisely enough to lift the value of the firm and thus generate capital gains. Personal income excludes capital gains because national income statistics measure flows of income from current production, not asset sales. That is also true of GDP, adding another reason to discard GDP as the basis of comparison.

However, Congressional Budget Office reports on the distribution of income do include realized capital gains when assets are sold (turning wealth into income). 

CBO Shares of Household Income

The second graph shows that labor’s share of household income is highest in deep recessions (77.5% in 1982, 76.2% in 2009) and lowest at cyclical peaks (70.6% in 2000, 68.3% in 2007). The higher labor share in recessions does not mean recessions are good for workers, of course, but that they are even worse for business and investors. Those who equate a higher labor share of income (e.g., during recessions) with higher real income for workers are making a basic and very large mistake. 

Capital income was highest in the early 1980s because the Federal Reserve kept interest rates very high, and capital income (dividends, interest, and rent) has shown no upward trend since then. Dividends and rent are up, but interest income is down.

Capital gains rose at specific times, but there has been no upward trend. There was a spike in capital gains in 1986 because the tax on gains jumped to 28% the following year. Realized gains also rose for four years after the capital gains tax was brought back down to 20% in 1997, and again after the capital gains tax was cut to 15% in mid-2003.

The white space at the top is important because it increases by four percentage points from 1990 (15.3%) to 2016 (20.3%) while labor’s share fell by 2.5 percentage points (from 75% to 72.5%). That white space is transfer payments: income from neither labor or capital. As the first graph showed, labor’s somewhat smaller share of income is not because of any sustained rise of capital income or capital gains. It is because of a sustained rise in the share of income from transfer payments and a sustained fall in the labor force participation rate.

Meanwhile, household income from owning a closely-held private business doubled since 1986: from 4% of household income in 1986 to 8% in 2013. That reflects the well-known shift of income from corporate to “pass-through” entities after 1986 as the top individual tax rate became even lower than the corporate tax rate (1988-92) or about the same dropped to the same as the corporate rate (35% 2003-2012)) or lower. That did not mean that “business” grabbed a bigger share at the expense of “labor,” but that a larger share of business income shifted from corporate to personal data.

The frequently repeated angst about “the fall of labor’s share of GDP in the United States” is based on a serious yet elementary misunderstanding of both labor income and GDP. “Labor’s share of GDP” is fundamentally nonsensical, because so much of GDP (depreciation, defense, etc.) could not possibly be paid to workers, and because the measure of labor income is too narrow (excluding the self-employed). 

Labor’s share of the CBO’s broadly-defined household income also fell (unevenly) because the share devoted to transfers rose, but also because the share moved from corporate to household accounts (and individual tax returns) also rose. Business income counted within CBO’s household income has increased its share of such income since the Tax Reform Act of 1986, but that just reflects a change in organizational form from C-Corporation to pass-through status.

Labor’s share of personal income fell mainly because the share devoted to government transfer payments rose. Labor’s share of GDP fell for other reasons (rising shares going to housing, government, and depreciation), but it is a fundamentally misconstrued statistic used to rationalize irresponsible remedies to an illusory problem of “monopolies.”

 

 

October 26, 2017 9:00AM

It’s Time for the Supreme Court to Take Up the Second Amendment Again

By Ilya Shapiro and Matthew Larosiere

Because the Second Amendment protects the right to bear all arms in common lawful use, any law that limits that right has to pass certain standards to be constitutionally permissible. The courts haven’t yet ironed out exactly what kind of scrutiny laws implicating the Second Amendment must pass, but such laws must have an important government objective and not be so broad as to burden protected activities unrelated to the harm they’re designed to address.

California requires most firearm purchasers to wait 10 days before they can bring their gun home, regardless of whether they already own one, or how long it takes to pass a background check. Several California residents who already own firearms challenged the 10-day waiting period and prevailed in federal district court because the state could only assert a general interest in a “cooling off” period.

The U.S. Court of Appeals for the Ninth Circuit ignored that the burden was on California to prove its case—and the state could show no evidence that the wait would have any public-safety effect when the purchaser already owns other arms. Instead, the court speculated as to what kind of harms the law might conceivably prevent, not any important interest it does actually serve. In the face of a decision that would allow California to arbitrarily infringe their Second Amendment rights, the challengers now seek Supreme Court review.

Cato has filed a brief supporting their petition. Silvester v. Becerra is an important case that the Court should not let slip by, because it presents an opportunity to provide much needed guidance in an area where lower courts have failed to reach a consensus on anything from what is protected by the right to keep and bear arms to the appropriate level of scrutiny judges should apply when considering lawsuits involving Second Amendment rights.

The morass of case law that has developed since District of Columbia v. Heller (2008) and McDonald v. City of Chicago (2010) is so divergent that the opinions of individual circuits read as though there was no precedent in this area whatsoever. From the Fourth Circuit’s holding that common semi-automatic weapons are “beyond the scope” of constitutional protection, to the Second Circuit’s deciding that only “severe” restrictions of the right to bear arms warrant any form of heightened scrutiny, all that’s clear is that the Supreme Court needs to clarify what hurdles the government must cross to justify violating what it itself held in Heller is a fundamental right.

The case here is quite narrow, covering only the application of an arbitrary waiting period to people who already own guns. The Court’s input, then, would not upset the diverse tapestry of gun laws that have developed across the country. Instead, it could help enable lower courts to competently move forward in developing Second Amendment jurisprudence. Because the case is small, the facts straightforward, the error below so clear, and the issue so constitutionally significant, the Supreme Court should step in and remind the Ninth Circuit what was said in McDonald, that the Second Amendment is not a “second-class” right for the circuit courts to “single out for special—and specially unfavorable—treatment.”