Thinking Through the Loma Prieta Lesson

Thirty-five years ago this evening, I fell over at soccer practice during a routine drill—then ducked and covered, the standard earthquake protocol California’s public school system taught. (When the earth shakes violently, one doesn’t stop and think rationally about whether anything can actually fall on them in the middle of an open field.) These eight seconds may seem distant and trivial now. But they can yield useful insight for investors weighing the costs—and risks—of natural disasters today.

Any time disaster strikes, we hear about the rising costs, which headlines usually link to climate change. The implication is that damage costs and economic losses have soared because disasters are bigger and more monstrous in this day and age. Never mind the science, politics and debates behind all of this, nor the violent hurricanes that struck during Victorian times. Nor is this about the infrastructure decisions helping increase flooding in some areas over time. Important and interesting as it all is, it is sociology, outside the direct economic factors stocks care most about.

Instead, let us explore development—a factor many discussions fail to account for when comparing disaster costs across eras. This is an absolute must when weighing market risks. And on its 35^th anniversary, the Loma Prieta earthquake will be our guide.

Loma Prieta was a beast. Though hitting “only” 7.1 on the Richter Scale, it caused devastating damage across several Bay Area counties. This is the quake that wiped out much of downtown Santa Cruz, leveled a swath of San Francisco’s Marina District, disrupted the World Series and collapsed major double-decker freeways (including a portion of the Bay Bridge). We were without power for days and rationing transistor radio batteries, but as news of the devastation gradually trickled in through family friends, it felt massive and unreal.

The physical damage’s cost was also only $5.9 billion, which was 0.1% of US GDP at the time.^[i] Adjusted for inflation, it would be $15.3 billion in today’s dollars. Next to the massive damage costs of recent hurricanes, it is a pea.

But hold on. Today, the damage would probably be far costlier, for a simple reason: The Bay Area changed. A lot. In 1989, Silicon Valley wasn’t the Silicon Valley you think of. Of today’s Tech giants, only Apple was headquartered here, and it was not in a golden age. Chip makers and defense contractors comprised the bulk of the corporate Tech scene, surrounded by a sea of struggling startups in low-slung office parks. There wasn’t a lot of money sloshing around. The area was affordable enough for single-income families to live comfortably, even when that single earner was a teacher, firefighter or police officer. A three-bedroom suburban ranch house went for about $300,000, give or take, depending on the square footage and school district. There were still cherry orchards, farms and wide-open spaces.

These days, there is a heck of a lot more physical capital, and it is all vastly more expensive. House prices have soared to absurd levels, bringing those same starter ranch houses to about $3 million and counting. The open spaces are now major Tech campuses populated by names you are very familiar with. So is much of the bayshore, all built on silt and landfill. The farms and sleepy strip malls are now multistory luxury apartment blocks. The cherry orchards are big shopping plazas. Suburban downtowns are no longer small train stops and main streets with struggling boutiques and bad restaurants. They are big. Built up. Sprawl. Downtown San Francisco, mostly old and Art Deco or Brutalist 35 years ago, is now doing its best impression of downtown Dallas, with soaring glass towers.

Yes, many buildings have gone through seismic retrofits and construction standards have improved, but soil liquefaction is still an issue in the low-lying areas. These structures whose value has skyrocketed are 35 years older, with the attendant wear, tear and decay. So while I can’t estimate the costs, it seems fair to presume a Loma Prieta-sized quake with a similar epicenter would leave far, far bigger dollar signs in its wake for the simple reason that a lot more and more valuable property would be in its path.

Aside from my anecdotal observation of house prices, the above is all qualitative. So here are some numbers. In 1989, the combined gross regional product (GRP) of the affected counties was $174 billion.^[ii] For those curious, this includes Alameda, Contra Costa, Marin, Monterey, San Benito, San Francisco, San Mateo, Santa Clara, Santa Cruz and Solano counties. In 2022, the latest county-level data available, these counties combined to GRP just over $1.2 trillion.^[iii] That is a leap from 24.1% of California GDP in 1989 to 33.4%.^[iv] As Tech boomed, the region vastly outgrew the state and nation.

And yet, we still haven’t scratched the surface of the actual economic effect then or now. Property losses get quoted the most in disaster coverage, but that isn’t synonymous with the economic hit. Property losses affect the value of a region’s physical assets—a stock measure. GDP is the annual (or quarterly, monthly, whatever) flow of activity, the sum of spending and investment. Cruelly, property loss eventually counts as a positive toward GDP once rebuilding begins and governments, businesses and people spend on replacement and reconstruction. THIS IS NOT STIMULUS. It is a redirection of money that, absent the destruction, would have been spent or invested elsewhere. But GDP math counts it as positive.

The main negative effect on GDP wouldn’t come from structure losses, but from the commerce that doesn’t happen as a result of the disaster—business lost due to power outages, damaged buildings, roadways and that sort of thing. Bay Area government researchers estimate the 10 affected counties suffered between $725 million and $2.9 billion in lost output during the month or two businesses were digging out and getting back online, with about 80% recovered the following quarter.^[v] They further estimated a “permanent loss”—business that left and never came back—of $181 million to $725 million.^[vi] For households, the lost wages and salaries amounted to about $54 million.^[vii]

This is a lot smaller than the property damage tally. Researchers note it probably would have been worse if Loma Prieta had affected more than the Bay Bridge, the Embarcadero Freeway and the Nimitz Freeway’s Cypress Structure. They note the Bay Bridge’s closure caused a temporary decline in San Francisco’s sales tax revenue and posit that, if more freeways were gummed up, regional activity would have been harder-hit.

All interesting, but we can’t use this to guess at the economic effect today. The lost activity could be higher, relative to state and national GDP, because of how much more this region produces. Or it could be lower because we have this thing called the Internet, enabling households to purchase online what they can’t buy from a closed store—and enabling office workers to log in remotely. Even in 1989, most of the economic losses concentrated in retail. Everything else went back online pretty quickly (including, much to my chagrin, the schools).

So are disasters’ actual economic effects getting smaller or bigger? The frustrating answer is that it depends … and is exceedingly difficult to estimate. Insured losses are getting bigger, but that is largely because there are more and more valuable structures and businesses in the affected areas. Loma Prieta’s insured losses were about $900 million.^[viii] They would be much higher now, most likely, due to the physical development and home prices’ exponential rise. We have seen this in hurricane-affected cities as well. The sun belt has grown hand over fist, putting more buildings in storms’ path while driving up property values. The mounting costs that steal headlines are, to a large degree, a side effect of rip-roaring growth.

But for stocks, it also tends to be way beside the point. We know, from US economic history, that whatever commerce goes offline temporarily or permanently, it tends to barely rate in national and global output. This renders the associated hit to corporate earnings tiny, which is why stock markets tend to overlook these things and move on quickly. Andrew, Katrina, Ike, Harvey, Sandy and all the rest—tragic as they were—didn’t cause recessions or falling stock markets. Nor did Loma Prieta, Northridge or the other monster quakes. The major exception, Japan’s Great Tohoku Earthquake and tsunami in 2011, did have a measurable effect—but mostly because Japan is a smaller geography and economy, as well as the associated nuclear disaster, which hammered a region and affected the country’s power supply longer-term.

Overall, then, investors seem best off approaching natural disasters as markets do: Sift the headlines from reality. Note the incorrect conflation of property damage and economic effect. Remember the effects are local and isolated, representing a tiny sliver of corporate revenues and earnings over a finite period. That doesn’t negate the human impacts, which are real and tragic. But markets deal with cold, hard facts.