A Long Game

“Every season since the Gold Rush, California has blossomed with new money — first in gold, then in land, cattle, railroads, agriculture, film images, shipbuilding, aerospace, electronics, television and commercial religions. The ease with which the happy few become suddenly rich lends credence to the belief in magical transformation.” — Lewis Lapham, 1979, “Lost Horizon”

“When do you think people in the Bay Area started to realize that you could make more money from tech than from real estate?” Jed Kolko asked me.

We were sitting at Ma-velous, a coffee shop frequented by San Francisco’s political movers two blocks from City Hall and kitty corner from Twitter’s headquarters in the Shorenstein-owned former San Francisco Furniture Mart.

Just outside the window was Market Street, San Francisco’s main thoroughfare. In 1847, two years before the Gold Rush transformed the city into a teeming boomtown, a 30-year-old Irish immigrant named Jasper O’Farrell presciently surveyed the street to be 120 feet wide  — which is broader than the Philadelphia main street it was named after  — even though the city was only 500 residents strong at the time.

“1980s,” I said in response to his question. Seemed obvious. The PC revolution. Steve Jobs and Apple. And a decade after the Peninsula was christened “Silicon Valley” by an electronics trade newsletter and the “Fairchildren,” or alumni of Fairchild Semiconductor, went on to found Intel, National Semiconductor and AMD.

2005 was his guess. The year after Google IPO-ed, and when escalating home prices made real estate an ever-riskier bet. That was the year that he felt that more people started coming to San Francisco for professional ambition rather than lifestyle reasons such as being openly gay or for pursuing callings that didn’t need to be so well-compensated.

Like a handful of other people at the intersection of real estate and technology, Kolko would be in a position to make an educated stab. He led some research at the well-regarded nonprofit, non-partisan think tank Public Policy Institute of California for five years, and then was the head economist at real estate startup Trulia.

It was a provocative rhetorical question. Land, labor, capital. Technological change. How do they intersect?

Cities, or physical communities, are one of the hardest kinds of scaling problems that exist.

In some ways, we’re lucky that the first two decades involving the advent of the commercial Internet were largely a positive-sum game. The creation of digital space for self-expression, at near-zero cost, does not necessarily challenge or erode someone else’s right to space or resources.

That makes it easy to forget that California is a state built throughout generations of conflict over land and its inherent constraints. It is this oldest of resources  — not capital or human talent or ingenuity  — that most constrains the region’s potential for workers at all income levels.

Cities, or physical communities, are one of the hardest kinds of scaling problems that exist. The infrastructure needed to sustain bigger networks, such as schools, sewers and mass transit, gets harder  — not easier —  to develop, finance and maintain the larger your population becomes.

There are countless constituencies of different incomes, racial groups, interests and professions that are all vying with each other for limited space. Although it’s possible to build up, urban development is much more of a zero-sum game.

With every major economic shift ,  from an agrarian to an industrial economy, and then an industrial to a knowledge-and-services economy , California and the United States have faced distinctive junctures in their approach to land-use and housing.

California’s fragmented, post-war suburban model, which was created for a more even wage distribution in a mass industrial economy, is clearly becoming more dysfunctional by the year.

I believe we’re hitting another major juncture, although I don’t know when it will deteriorate to the point that it forces real reform. California’s fragmented, post-war suburban model, which was created for a more even wage distribution in a mass industrial economy, is clearly becoming more dysfunctional by the year for a knowledge-and-services economy with a wider level of income stratification.

Not only are we not building enough housing overall, we have scarce sources of funding for supporting those on the lower-earning ends of a rapidly widening income spectrum. So we end up politicizing and extracting funds out of new construction even though we are 40 years deep into a largely self-imposed housing shortage.

There are a couple of disturbing trends showing up in the data. If you look across the state’s workforce, Californians born in 1990 are on average spending 50 percent of their income on housing. That’s way above the 30-percent-of-income level that is generally considered to be the threshold of whether housing is affordable or not in public policy conversations.

income-spending-housing

Then, if you look at working-class segments, commutes are rapidly rising for the lower-income workers in the region:

This is troubling because commute time is one of the strongest predictive factors in determining a child’s chances of climbing from the lowest income quintile to the highest-earning one. That morning and evening time between parents and children that is taken up by commuting is invaluable for bonding and child development.

Cities around the Bay Area are starting to contort themselves into stranger and stranger positions just to support basic public services.

Cities around the Bay Area are starting to contort themselves into stranger and stranger positions just to support basic public services. Suburbs like Cupertino, where Apple is headquartered, are now having to build teacher public housing projects for tens of millions of dollars because the cost of living is too high for an entry-level teacher making $55,000 a year.

Meanwhile, the city has approved another Apple campus that will bring 13,000 additional Apple employees to the city, while only committing to building 1,400 housing units over the next seven years.

I’m not sure when a breaking point happens, but I want to offer some essays and short pieces over the next few days to give you a couple of takeaways. The TL;DR is that I’m going to start working on new projects soon, but I want to leave a map behind of what I think has to be done long-term in Northern California.

Here are the additional pieces (and I’ll be posting more in the days to come):

An old problem

California has faced housing and land shortages multiple times and has changed its regulatory regimes in response. I’m going to start with two histories from the state’s first Gilded Age and post-war era.

In the first Gilded Age, the concern was over land monopolization by a handful of large-scale owners and how that crowded out and impoverished labor. In the postwar period, California leveraged the automobile and federal subsidies to unlock previously inaccessible land and create a golden age of cheap housing and suburbanization.

This boom period ended in the 1970s as the state’s flat, developable coastal lands were built out and the oil crisis made sprawl more expensive. That’s when housing shifted from being perceived as a consumable good to an investable asset.

Since then, a new generational land cartel has emerged with Californian Baby Boomers protecting entitlements and higher property values for themselves in the form of land-use restrictions and Proposition 13. Global capital has been subverting and taking advantage of these favorable legal and taxation protections on real estate in a extremely low interest-rate world. All of this has come at the cost of the state’s working and middle class and its future workforce.

It’s a global issue

It faces major cities in all economies that chose a housing-as-an-investable-asset model following World War II like the U.K. and Australia.

After the credit-fueled housing crises of the 2000s tested the upper bound of what the homeownership rate could be, many countries are grappling with lower homeownership rates for the foreseeable future. This raises questions about what the future balance of tenant and homeowner rights and subsidies should look like.

There are alternate approaches in industrialized countries like Germany and Japan that make housing more of a consumable good rather than an appreciable asset, but it’s hard to see how we could ever shift toward those models since there is so much national wealth tied up in housing.

California is mostly half-assing solutions

There is no “magical transformation” here (unless VR and telepresence renders location obsolete, which the Internet certainly didn’t do). Technology companies and political leaders completely underestimate the depth of reform that’s needed to actually solve housing affordability.

Right now, San Francisco and New York City are reliant on inclusionary zoning, which effectively taxes the production of new housing in order to finance subsidized, affordable housing. It’s a Band-Aid solution that lets everyone stay in a system structured to keep property values higher (e.g. making housing less affordable) while providing a teeny-tiny, token amount of low-income housing.

Dumping hundreds of millions of dollars of public or philanthropic capital down a badly designed property system will not get you very far.

For example, inclusionary housing has produced 1,787 below-market-rate units and $59 million in affordable housing fees in San Francisco since 1993. Meanwhile, the value of the city’s assessed property rose by $11 billion in the last year. We need to capture more of the land value increases that property owners are accruing through no particular special effort of their own, perhaps in the form of a deferred transfer tax at sale. Unfortunately, most jurisdictions in the state are designed to be majority homeowner, and the state’s Third Rail, Proposition 13, hasn’t been touched since the late 1970s. It’s hard to see a political constituency that could change this unless it gets really bad.

That said, California’s homeownership rate is the lowest it’s been since 1991 and has dropped a full 6 percentage points since the financial crisis to 54.2 percent. Maybe we just have to wait another 10 to 15 years until the composition of the electorate changes enough for reform.

It has to do with the way we deal with land.

While the technology industry and its compensation practices have increased the region’s income inequality over the last generation, taxing businesses, wealth or investments alone for affordable housing funds won’t solve the housing issue.

Dumping hundreds of millions of dollars of public or philanthropic capital down a badly designed property system will not get you very far because it will drive land values even higher, producing an unearned windfall for property owners.

Even though developers tend to get pilloried in the public process, it is the land owners that get egged on by local brokers to sell at the absolute highest price.

For San Francisco to build affordable housing in the Mission District today, private land owners are demanding $250,000 per housing unit in just land costs from the San Francisco city government because they can. The Latino community fought and used the threat of a moratorium on new housing construction to wrestle $50 million out of the $310 million bond that voters approved last fall. It was the first affordable housing bond the city had passed in almost 20 years. But then the Mission community realized that $50 million gets you — drum roll — three plots of land!

Keep in mind that California also already has the most progressive income taxation system of any state in the U.S. because it also has some of the lowest property tax rates, especially on properties worth more than $1 million. The downside of being extremely reliant on income and capital gains taxes and having low property taxes is that the California state government’s revenue structure is highly volatile and predictably goes into steep deficits during recessions.

So while you could go down this route — and there are interesting conversations about progressive equity, tech executive compensation and carried-interest loopholes — you’d have to do it in addition to land-use and property taxation reform to have any real impact on housing affordability.

An economic downturn won’t bring lasting affordability

An economic downturn may soften pressure on local housing markets, but it will not fix this problem. The Bay Area’s housing market tends to rise and then plateau; this is a function of how local land markets work.

Even though developers tend to get pilloried in the public process, it is the land owners that get egged on by local brokers to sell at the absolute highest price. Because there is no structural disincentive to sitting on underutilized land because of property tax caps, land owners can just sit a weak cycle out, withhold their land from the market and wait for the next upswing.

While there are quasi-public institutions that sustain mortgage and home-buying demand through weak parts of the economic cycle, there isn’t really an equivalent on the construction financing side. When markets turned in 2008, project financing evaporated, which left the city without a decent construction pipeline until it was too late in the up-cycle. The same thing may happen in the next downturn.

Beyond the Bay

If the tech industry wants a faster solution than waiting for several decades of reform, it could decentralize and encourage more viable tech hubs elsewhere.

When Detroit’s automobile industry was around the same age that Silicon Valley is today, it began decentralizing production in the 1940s, decades before the 1967 race riots and competition from Asian automakers.

However, even if tech decentralizes a little bit, the state’s working- and middle-class are still screwed because these are issues that are deeply rooted in California’s approach to land-use and property taxation that stretch back decades. They’re visible all over the state well outside of Silicon Valley.