Ownership Transparency Works: Geographic Targeting Orders in the US Real Estate Market

The anticorruption community, along with those concerned about tax evasion, fraud, and other forms of illicit activity, has made anonymous company reform a high priority on the reform agenda. It’s not hard to see why: Kleptocrats and their cronies, as well as other organized criminal groups, need to find ways to hide and launder their assets, and to do so in ways that are difficult for law enforcement authorities to trace. Moreover, those whose legitimate sources of income would be insufficient to obtain luxury assets would like to conceal their ownership of such assets, as the ownership itself could arouse suspicion, and might make the assets more vulnerable to forfeiture.

So-called “know-your-customer” (KYC) laws in the financial sector have made it much more difficult—though, alas, far from impossible—for account owners to conceal their identities from the banks and government overseers, at least in the US and most other OECD countries. But it is still far too easy for criminals to purchase substantial assets in wealthy countries like the United States while keeping their identities hidden. All the bad actor needs to do is, first, form a company in a jurisdiction that does not require the true owner of the company to be disclosed and verified to the government authorities, and, second, have this anonymous shell company purchase assets in a transaction that is not covered by KYC laws. Step one is, alas, still far too easy. Though we often associate the formation of these sorts of anonymous shell companies with “offshore” jurisdictions like the British Virgin Islands, in fact one can easily form an anonymous shell company in the United States. Step two, having the anonymous company purchase substantial assets without having to disclose the company’s owner, is a bit trickier, because you’d need to avoid the banking system. But you can get around this problem by having your anonymous company purchase assets with cash (or cash equivalents, like money orders or wire transfers), so long as no party to the transaction is under obligations, similar to those imposed on banks, to verify the company’s true owner.

One of the sectors where we’ve long had good reason to suspect this sort of abuse is common is real estate, especially high-end real estate. Though money laundering experts had long been aware of the problem, the issue got a boost from some great investigative journalism by the New York Times back in 2015. The NYT reporters managed to trace (with great effort, ingenuity, and patience) the true owners of luxury condos in one Manhattan building (the Time Warner Center), and found that a number of units were owned by shady characters who had attempted to conceal their identities by having shell companies make the purchases.

The US still hasn’t managed to pass legislation requiring verification of a company’s true owners as a condition of incorporation, which would be the most comprehensive solution to the anonymous company problem. Nor has the US taken the logical step of extending KYC laws to real estate agents across the board. But starting back in 2016, the US Treasury Department’s Financial Crimes Enforcement Network (known as FinCEN) took an important step toward cracking down on anonymous purchases of luxury real estate by issuing so-called Geographic Targeting Orders (GTOs). And thanks to some excellent research by the economists C. Sean Hundtofte and Ville Rantala (still unpublished but available in working paper form), we have strong evidence that many purchasers in the luxury real estate market have a strong interest in concealing their true identities, and that requiring verification of a company’s ultimate beneficial owners has a stunningly large negative effect on the frequency and aggregate magnitude of anonymous cash purchases.

Before proceeding to Hundtofte & Rentala’s main findings, some quick background on FinCEN’s GTOs. In a nutshell, these GTOs require that, for covered real estate transactions, the title insurance company that provides coverage to the purchaser in an all-cash transaction by a company must verify the ultimate beneficial owner (defined as a person who owns, directly or indirectly, at least 25% of the equity in the company), and provide that information to the Treasury within 30 days after closing. (The regulation applies to the title insurance company, rather than the seller and/or buyer, because FinCEN has the legal authority to impose regulations on title insurance companies but not to the other parties to the transaction. It’s possible to purchase real estate without title insurance, which would allow the buyer to remain anonymous, but doing so is very risky.) Whether a transaction is covered by a GTO depends on the location of the property and the purchase price. The first GTOs, announced in January 2016 and implemented in March 2016, only covered two markets: Manhattan (for purchases over $3 million) and Miami-Dade County (for purchases over $1 million). In July 2016, FinCEN announced that it was expanding the program to cover a number of other major metropolitan areas—including Los Angeles, San Diego, San Antonio, San Francisco, Palm Beach, and the rest of New York City—with purchase price thresholds ranging from $500K to $2 million; these GTOs became effective in August 2016. In August 2017, FinCEN announced an additional GTO for Honolulu. And most recently, in November 2018, FinCEN further expanded the program with an additional set of GTOs to cover a number of other cities, including Boston, Chicago, Dallas, Las Vegas, and Seattle, and lowered the purchase price threshold for triggering the ownership verification requirement to $300,000 in all of the covered markets. Penalties for violating the GTOs are stiff, with fines up to $250K and five years in prison for willful violations.

If it’s in fact true that a desire to maintain anonymity is a major reason for all-cash real estate purchases by companies, then the GTOs—if effective—should substantially reduce such purchases in covered markets. If, on the other hand, anonymity is merely incidental to, and not the main reason for, the bulk of these transactions, then the GTOs shouldn’t have much of an effect, since the ownership verification would be a minor formality in cases where there was some other good reason to structure the purchase as an all-cash purchase by a company. We would also see little effect on the proportion of all-cash company purchases if the GTOs were easy to evade.

Hundtofte & Rentala set out to find out what impact the GTOs had, and the effects were dramatic and unambiguous: The introduction of the GTOs caused all-cash purchases by anonymous companies to dry up. They only have data covering 17 states (plus DC), and their data only goes up to the end of 2016 (meaning that they can’t measure the impact of the expansions of the GTOs in 2017 and 2018), but the data that they do have tells a clear story. They do all the requisite fancy statistical analysis (regression discontinuity analysis, as well as differences-in-differences analysis using counties not covered by the GTOs as controls), but really all a lay reader needs to see to understand the main punchline are the following two pictures.

First, here’s a graph showing the weekly dollar value of real estate transactions over the 2015-2016 period. The solid line represents cash purchases by companies; the dashed line represents all other purchases (purchases with a mortgage, by an individual, or both). The scales are different, but the purchase volumes move in lockstep right up until the first GTO goes into effect in March 2016, after which point all-cash purchases by companies crater:

And here’s a second picture, limited to Miami-Dade County. (Apologies but for some reason the scale and axis labeling gets screwed up when I copy the figure—you can find the original on p. 27 of the working paper. The time scale on the x-axis is the same, and the vertical gray lines again represent the dates for GTO announcement and implementation.) Here for some reason the authors have reversed the coding such that the dashed line represents cash purchases by companies and the solid line represents all other purchases, but the bottom-line result is basically the same: Cash purchases by companies were a significant part of the Miami real estate market before the Miami GTO went into effect in March 2016, and after that date such purchases dried up almost entirely.

Those pictures tell most of the story. The more comprehensive statistical analysis fleshes this out a bit more. Among other things, the authors estimate that prior to the GTOs, cash purchases by anonymous companies made up roughly 10% of the dollar value in residential real estate purchases, and after the GTOs were introduced, all-cash purchases by companies dropped by roughly 70%. Interestingly, the announcement of the original GTOs for Manhattan and Miami appeared to have had a big effect in other markets, to which GTOs only extended later—perhaps, the authors speculate, because FinCEN made it clear that expansion was coming, or perhaps because the GTOs signaled more generally that the US government was going to be scrutinizing real estate purchases more closely. The study also found that, following the introduction of the GTOs, the overall price of luxury real estate in the covered markets declined by a substantial amount—around 4-5%–suggesting that a significant demand for anonymously-held luxury assets was driving up prices.

What do we learn from this? A few things. First, and perhaps unsurprisingly, there’s a big market for secrecy. The anonymity that comes with companies rather than individuals purchasing high-end real estate is not merely incidental, but rather—in the large majority of cases—the main reason that the companies rather than the individuals are the formal purchasers. That does not by itself demonstrate that most of these purchases were to facilitate money laundering or other criminal activity, but it certainly suggests that we’re right to be worried that many purchasers are trying to hide their identities. Second, both the magnitude and speed of the GTOs’ impact demonstrates how relatively simple transparency reforms can have a big and immediate effect. This doesn’t mean that we’ve solved the problem—it’s quite possible, indeed likely, that bad actors will shift from luxury real estate purchases to one of the many other loopholes through which they can buy assets without having to reveal their identities to regulators, or perhaps to real estate purchases in other jurisdictions. But it does suggest that simply requiring the provision of beneficial ownership information in high-risk transactions can have a strong deterrent effect. And this in turn suggests that requiring the provision of beneficial ownership information at the moment of company formation—a reform that the US Congress will consider again this year—is an idea whose time has come.

7 thoughts on “Ownership Transparency Works: Geographic Targeting Orders in the US Real Estate Market

  1. A slight coda to Matthew’s very fine piece from Franklin Foer’s somewhat distraught take on Trump’s influence on America, “Russian-Style Kleptocracy Is Infiltrating America,” in the current Atlantic (https://www.theatlantic.com/magazine/archive/2019/03/how-kleptocracy-came-to-america/580471/). While adding some useful background on why cash purchases of real estate have avoided scrutiny for so long, it overstates Trump’ influence on the career civil servants that enforce American law —

    “[N]estled in the [stringent antimony laundering provisions contained in the 2001] PATRIOT Act lay the handiwork of another industry’s lobbyists. Every House district in the country has real estate, and lobbyists for that business had pleaded for relief from the PATRIOT Act’s monitoring of dubious foreign transactions. They all but conjured up images of suburban moms staking FOR SALE signs on lawns, ill-equipped to vet every buyer. And they persuaded Congress to grant the industry a temporary exemption from having to enforce the new law.

    “The exemption was a gaping loophole—and an extraordinary growth opportunity for high-end real estate. For all the new fastidiousness of the financial system, foreigners could still buy penthouse apartments or mansions anonymously and with ease, by hiding behind shell companies set up in states such as Delaware and Nevada. Those states, along with a few others, had turned the registration of shell companies into a hugely lucrative racket—and it was stunningly simple to arrange such a Potemkin front on behalf of a dictator, a drug dealer, or an oligarch.
    . . .
    “Over time, the gap between the noble intentions of the PATRIOT Act and the dirty reality of the property market became too wide to ignore. In 2016, Barack Obama’s administration tested a program to bring the real-estate industry in line with the banks, compelling brokers to report foreign buyers, too. The ongoing program, piloted in Miami and Manhattan, could have become the scaffolding for a truly robust enforcement regime. But then the American presidency turned over, and a landlord came to power. Obama’s successor liked selling condos to anonymous foreign buyers—and may have grown dependent on their cash.”

  2. While GTOs seem very promising, I can’t help but wonder why Congress has declined to implement what the post acknowledges is a much simpler solution––a nationwide law requiring verification of a company’s owners. Does blame lie solely with lobbying efforts by the real estate industry, as argued by the Atlantic article Rick cites? Are there substantive arguments against such a law? None come to my mind, at first glance.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.