How I'm voting, Part 2: CA Voter-Nominated Offices, Oakland Initiatives (2018)

October 14, 2018

The second in my series giving some opinions on California's election. For California's ballot initiatives, see part 1.

Part 2: California Voter-Nominated Offices

United States Senator: Feinstein

Policy-wise, it's a wash; both are party-line Democrats, pushing environmental issues, women's rights, and opposing Trump.

The question is, who do you want as your senator?

Feinstein is the incumbent; she's facing a challenge from the left by Kevin de Léon, a new Democrat of the AOC school. Like many of his type, he's pushing a populist agenda that emphasizes conflict, and unworkable policies like "Medicare-for-all", an unaffordable plan with zero chance of passage, given its politics.

I also admire how Feinstein conducts herself as a Senator. During the Kavanaugh confirmation, Sen. Harris was shouting and walking out; Feinstein let things run their course. Note that both got to the same place—a vote against—but Feinstein got there while upholding the dignity of the Senate, while Harris insisted on theatrics. My HOA work has given me an even greater appreciation of the need for civility in politics; it's rare.

Governor of California: Toss-up

This is a hard one.

Newsom is an artful dodger; a stuffed-shirt politician without substance. He says he supports "housing reform", but won't get specific on where he stands on controversial issues like Prop 13. His campaign reeks of arrogance and privilege—refusing the Cox campaign's requests for debate, declining to put a statement in the Voter Guide—the not-so-subtle message is, "We've got this thing locked up; on to bigger and better things".

I really want to vote for Cox, except he's a climate denier whose policies hew a little too close to Trump's. He also wants to roll back gas taxes, a step away from the settled economic consensus, and growing popular consensus, that we should tax carbon.

Realistically, Newsom's going to win; I think I'll vote for Cox just to show that victory is no sure thing.

Lieutenant Governor of California: Hernandez

Not much difference overall. I think Hernandez is slightly more qualified and care about his issues (chiefly immigration reform) more than hers. Hernandez is also more of a known quantity.

Chronicle endorsement of Hernandez

Secretary of State: Meuser

Objectively speaking, the CA Secretary of State is doing a poor job: inaccurate voter rolls, non-functional software, exorbitant filing fees. Have you been to the DMV recently? Should the minimum franchise tax be $800?

I also like that Meuser is well-read, as the link above shows. I don't like the nativist tone of his voter registration efforts but overall I think some change here could be a good thing.

Controller: Betty Yee

Most of what I've read (Chronicle, San Jose Mercury News) suggests Yee is qualified to do the job, and did a reasonable job in her first term.

Roditis's candidate statement is nonsense:

The Controller’s office doesn’t think you pay enough taxes. They want to tax your doctor’s visit, childcare, home and vehicle repairs, haircut, you name it they want to tax it. The result, you will pay hundreds if not thousands more a year in taxes. No on a Service and Labor Tax. We can’t afford it. As Controller, I will fight to make California affordable, not seeking new ways to tax you.

Roditis can't do a thing here. He's running for Controller, the executive agency responsible for policies set in the legislature. He doesn't set policy; he does what the legislature tells him to do.

Treasurer: Conlon

Both are qualified. Fiona Ma seems to have done a good job in her last term. Conlon wants to focus on unfunded pension and healthcare liabilities, a ticking time bomb that never gets the attention it deserves (look at Illinois to see how that story ends)

Attorney General: Bailey

As I've written before, the Bay Area has a serious problem with crime. Bailey seems like the candidate who will be a little tougher on crime, so he gets my vote. I don't want the AG wasting time "fighting predatory for-profit colleges that steal from our students"—the real crime is the cost of college in general, and if someone can figure out how to get rich educating people, all the better.

Insurance Commissioner: Poizner

Better-qualified (second term, great education), more specific proposals on what he'll do, and good track record of fighting insurance fraud. Also endorsed by the Chronicle.

He's running as an independent, which is baller.

District 2 Board of Equalization: No preference

Flip a coin. Malia Cohen is going to win because she's a Democrat, with her hard-left agenda of "people's interests before special interests", being a "fearless advocate of working people", and "championing the $15 minimum wage".

I would vote for Burns but he's a hardcore prop 13 supporter which I can't do. He also has zero chance of winning in California 13, a district with a D+40 PVI.

Superintendent of Instruction: Marshall Tuck

Public schools are an utter joke. OUSD (Oakland Unified School District) was plauged by an accounting scandal that led to a resignation earlier this year and now they're in hot water after being accused of violating Title IX after they cut a bunch of athletic programs. I don't know what it will take to get public schools back on track but I'd prefer we don't do more of the same.

The priority needs to be students and educational outcomes, not entrenched bureaucracy and unions; Tuck seems less in hock to them overall.

Part 3: Oakland Propositions

This is easy; most of these are absurd.

Measure V: Yes

Under Oakland's existing laws, medical cannabis companies face a 5 percent tax on their gross receipts, and those selling weed for recreation have to pay 10 percent. These tax rates are locked in by a previous ballot measure and can only be changed by voters.

Here we go again…

This seems reasonable but it's the textbook example of why the tax code shouldn't be set by ballot measure.

Measure W: Strong no

Housing shortage? Sure, let's impose a new tax of $3K-$6K when someone wants to keep their unit empty for whatever reason.

Yet another reason why we need a clean repeal of Prop 13; people paying $10K/year in property tax (which I am) don't let their place sit empty.

This is a punitive, outrageous new tax that does nothing other than punish owners unlucky enough to own rental property in Oakland. Just like CA Prop 10, stuff like this scares the hell out of those who'd otherwise invest in (aka build) new housing in Oakland.

There is no evidence this will do anything to alleviate homelessness or housing problems, but it's guaranteed to raise the cost of doing business, owning, or renting property in Oakland. There's no statutory requirement on how the estimated $200 million in new taxes will be used, which is even worse. This initiative is plainly about class warfare and punishing "the rich", without regard to whether the solution will work at all, let alone if it's fair.

Oakland already has every conceivable kind of tax, including sales, transient occupancy, real estate transfer, business gross receipts, and all kinds of parcel taxes; let's not add yet another, especially one so big that's not proven to solve anything?

Measure X: No

Leave the transfer tax alone. It's bad enough that Oakland charges 1.5% of value any time property is bought or sold; that's a $22.5K tax just for the privilege of buying or selling a $1.5 million property, on top of agent commissions, escrow fees, title insurance, and everything else. There's no need for this to be higher; advocates claim the tax should be "progressive"—it already is because smaller properties are taxed less (by virtue of the tax being pegged to value), we don't need to double it by making the rate lower on top of the lower sale amount.

Measure Y: No

It's hard enough to be a small property owner in Oakland. Don't make it harder.

Small property owners are the bread and butter of the low-end market. They are the very people who rent out the most affordable units to the lowest end of the market. Passing this measure will make it even more likely these people will exit the market entirely, adding to the problem of housing affordability.

The only solution is more supply, not more taxes, fees, and regulations.

Measure Z: Strong no

No way this belongs as a ballot measure; it's way too specific and technical.

This should be done through legislation.

Measure AA: No

Oakland's tax base is already $1.5 billion, up $500 million in the last 5 years. That's $7K/year for Caroline and I. I should be living in a country club for that kind of money.


I'll look at the Oakland candidates tomorrow.


How I'm voting, Part 1: CA Initiatives (2018)

October 13, 2018


My name is David Albrecht, I live in Oakland, CA and try to keep up with local news. This is my analysis of the issues.

A few things about me:

  • I consider myself a classical liberal, which means I'm both more conservative and more liberal, at the same time, than many Bay Area residents. However, I have friends across the entire political spectrum whose views I deeply respect.
  • I generally oppose ballot initiatives as a matter of principle; they're too hard to change and make compromise difficult. Some thoughts
  • I strongly reject identity politics; I care about policy, issues, and ideas more than who's in favor of, or opposed to, an idea
  • I consider myself a fiscal moderate; I think California could use less public spending, but would rather see a working government with fully-funded pensions than one that treats tax minimization as an explicit goal.

Having said that, here are my thoughts on the issues (first in a series).

Part I: California Propositions

Proposition 1: Yes

Narrow, simple initiative supporting a $4 billion housing bond, including $1.8 billion for multifamily, $450 million of transit-oriented development, and $1.0 billion for veterans.

Legislative analyst estimates the annual cashflow required to service this debt will be $170 million. Debt service occurs via the general fund, a $120 billion fund whose revenue derives overwhelmingly from income and sales (not property) tax; I prefer this because it spreads the obligation broadly across California taxpayers. A $170 million annual obligation expands the general fund by only 0.1%, which isn't much. Put differently, the bond's $170 million annual cost is only $4/capita given California's almost 40 million people. It will take a while to repay but it's a small cost to house veterans, farmworkers, and others using approaches endorsed by experts like SPUR.

It's also a very straightforward measure that raises money for a defined purpose and doesn't create long-term legal complexity, or entanglements.

Proposition 2: No

Too complicated; divided expert opinion.

The initiative purports to build housing for the homeless by diverting funds raised by 2004's Prop 63, to issue new housing bonds. The key issues seem to be whether diverting funds is legal, whether fees and interest will take too much money away from program expenses, and whether local (county) control is better than centralized state administration.

NAMI Contra Costa, Contra Costa's branch of the National Alliance on Mental Illness, opposes the measure, in part because it fails to address local legal barriers to supportive housing, including zoning that blocks construction of homeless shelters.

The California Budget & Policy Center has a good writeup, which failed to convince me this is good policy: Proposition 2: Should California Sell Bonds Backed by County Mental Health Funds to Develop Supportive Housing for Homeless Residents With Mental Illness?.

Proposition 3: No

Too big, too complex, questionable efficacy.

The proposition authorizes an $8.9 billion (!) bond issue, at annual cost of $430 million to the general fund, to pay for "water projects".

Per the voter's guide:

These funds fall into six broad categories, as summarized in Figure 1. Within these broad categories, the proposition includes around 100 subcategories for how certain amounts must be spent, including for particular regions of the state or on specific projects.

100 subcategories? That sounds like legislation.

From The Chronicle:

This scheme was devised as an initiative that is being funded, in part, by individuals and entities that are going to be receiving a share of the bond money. The pay-to-play aspect in itself should give voters ample reason to reject Prop. 3.

Proposition 4: No

Small, well-targeted, but better done via legislation.

Sidenote: all the hand-wringing about property taxes around this is idiotic; California's General Fund is not funded by property taxes, as anyone can find out with a 5-second Google search.

Proposition 5: Strong No

This is a terrible ballot initiative that's funded by realtors trying to get more people to buy and sell their houses (and thus generate more commissions).

Prop 13, one of the worst things ever to happen to California created a situation where those who have been in their houses a long time (20-30 years) pay a fraction of what new people pay in property taxes—as much as 80% to 90% less than what newer arrivals pay.

This initiative tries to "fix" things by permanently entrenching the tax break longtime residents get, by letting them take it with them, when they move. So even if they're in a $1 million house getting assessed as if it was a $100K house, if they move to a different $1 million place, they'll still get assessed as if they're in a $100K place.

It's absolutely nuts. Nowhere else does this and nothing about this is even remotely fair.

Please don't vote for this. It further entrenches longtime property owners at the expense of the young and new arrivals. It further entrenches a two-tiered tax system while ensuring the burden of funding local governments, including police protection, fire departments, roads, and other necessary municipal services are concentrated on the newest arrivals to an area. That is totally unfair, and is opposed by everyone from affordable housing advocates, to local service providers (nurses, firefighters, teachers).

This is one of the few ballot initiatives I consider unconscionably outrageous.

Proposition 6: No

California's constitution already encourages an unbalanced budget by requiring a simple majority to spend but a 2/3 supermajority to raise revenue (tax); this initiative makes it even worse by further handcuffing the legislature's power to tax.

Low taxes are good, but balanced budgets are even better. Passing this makes a repeat of the 2008 budget crisis more likely.

Proposition 7: No

A ballot initiative to change time zones? Is this a joke?

Pass a law.

Proposition 8: No

Massively complex initiative that requires detailed analysis of company operations; legislative analyst has no idea what it will cost to enforce, it will make healthcare harder to deliver, and it sets "revenue limits" to control healthcare costs? No way.

We need cheaper healthcare, but complex, hard-to-enforce price controls won't get us there—they'll just push suppliers out of the market. The ACA is a great example: the people who passed it talked a big game about healthcare costs, but since it's passed, the Wall Street Journal reported 50% of US counties have only one insurer in their exchange. Best-case scenario, this will make medical billing even more complex, and clinics will hire an army of "revenue analysts" to find the inevitable loopholes. In the end, the only people who win are the companies selling software and services to "optimize billing" (perhaps not such a bad business to get into…).

Proposition 9

This was Steve Draper's crazy plan to break up California. I would've loved to see a debate on this, but the courts shut it down.

Proposition 10: Strong no

The path to greater housing affordability is more construction, not more regulation and control.

Oakland vs. San Francisco is a good example. Oakland has remained affordable, and built, as San Francisco spirals further out of control, largely because of San Francisco's much more aggressive rent control.

Initiatives like this scare developers; would you want to build here if you can't be sure something crazy won't happen?

It's also complex. A clean repeal of Costa-Hawkins would be better, if that's what people want.

Proposition 11: No

Work rules in a ballot initiative?

At the same time, however, the measure requires that meal breaks (1) not be during the first or last hour of a shift, and (2) be spaced at least two hours apart. The measure requires ambulance companies to operate enough ambulances to meet these meal break schedules.

You want this carved into stone for the next 100 years, when we have autonomous ambulances, flying cars, and the job of being an EMT completely changes? Please, vote no.

Proposition 12: No

Again…animal confinement rules in a ballot initiative? We want to carve this into stone until it gets changed in another initiative?

This is insane.

That shouldn't even be in a statute (law), it should be decided by an executive authority like the FCC or FTC charged with enforcing a law. No way this should be decided by initiative.

Calf raised for veal: Must be able to turn around freely, lie down, stand up, and fully extend their limbs. 43 square feet of floor space required.

43 square feet…until we pass another initiative making it 44.


I'll do the candidates tomorrow. As promised, CA candidates and Oakland measures.


Marrying well

October 12, 2018

That's my father-in-law, Dr. Ed Souza. We don't live in the same town so a lot of our interaction happens over Twitter.

I've gotten to know the family I married into better since the wedding, and I really lucked out. Patrick is a great friend and coworker, and Caroline's parents are pretty great, too. I'm sure we'll fight a little but on the whole I feel I ended up with much better than I deserve.

Which got me thinking—"marrying well"—it's a phrase more traditionally associated with marriage to a man, connoting socioeconomic standing, class, and privilege.

So I can't help wondering: what is "marrying well" when one marries a woman?

I haven't fully worked it out yet, but it seems family is a big part. And maybe men can get rich through marriage, too: after all, that's part of the Goldman Sachs founding story: Samuel Sachs, partner to his father-in-law, Marcus Goldman.


Preferred and common are not the same

October 11, 2018

Preferred and common stock are not the same thing.

When investors buy shares in an early-stage company, they typically negotiate protective provisions, intended to limit their downside. I am not going to get into why these provisions are necessary, or whether they're fair—opinions range from "must-have" to "tool of oppression"—the point is that most venture term sheets have them. That's no accident; they're there because the protections they provide are valuable.

Common shares don't have these protections; common shares end up buried behind huge liquidation preferences without the pro rata, drag-along, or information rights typical of preferred, and definitely without the ratchets.

I usually treat common as worth 1/3 to 1/2 as much as preferred. The gap is bigger very early on, when there's a lot of risk, or a huge liquidation preference. It gets smaller as things stabilize and the company's value (hopefully) grows well beyond any liquidation preference.

Surprisingly, when people talk about this stuff, the difference between common and preferred often gets lost in the noise. That's a huge mistake.

Situation #1 where I see the distinction lost: reporting about new financing. Say a company has 1 million shares outstanding and investors agree to purchase 100,000 new shares. The shares have a bunch of protective provisions and investors pay $5/share. The press often imputes the preferred share value to the entire business, calling this a "$5 million company"—wrong. Fair-market value of common is only 1/2 to 1/3 as much as preferred, maybe $2/share. So the true value of the business is closer to $2 million than $5 million (quite the difference).

Another way this gets twisted: "we're giving you a $150K equity grant over four years". OK, 30K shares, but at $5—is that $5 the preferred price, or the common? Because the grant is probably on common. "Don't worry about it, they're the same". No they aren't. You're getting the basic model (common) but paying the premium price. That isn't right.

Don't fall for this stuff. Get educated and understand the difference.


VC Firm Names

October 10, 2018

Why is it that VC firms (and a lot of private equity firms, for that matter) love the following naming schemes?

  • A piece of the outdoors or nature (Charles River Ventures, Sequoia Capital, Shasta Ventures, Union Square Ventures, Bowery Capital, Oaktree)
  • Something classical or Latin-sounding (Opus Ventures, Clarium Capital, Amicus Capital)
  • A bunch of guys' names (KPCB, Blumberg, ESL Investments)

Stratton Oakmont (names) wasn't too far off.


Deep Space 大厦

October 09, 2018

I've been watching Deep Space 9 and can't help but think they should really rename it "Deep Space Condominium". Any group living situation will be at least a little bit like DS9 (a show about life on a space station), but it's especially relevant to my building, in Oakland Chinatown:

  • Both have understaffed maintenance teams working through huge deferred maintenance backlogs, and residents who dislike waiting their turn
  • Both feature several cultures living in close proximity who think each others' food is gross and generally don't trust each other
  • Both have tons of rumors and gossip
  • Life at both has had its ups and downs, but the new administrations in both places are headed in the right direction

(The title of this post comes from the name of my building, 中山大厦)


No on everything

October 08, 2018

I have a standing policy of voting "No" on almost every California ballot initiative. I do this for everything—state, county, city. I have two reasons.

For one, ballot initiatives are too hard to change. Prop 13 capped California property taxes at 1% [1] in the late '70s because of Prop 13, and there's not a thing the legislature can do about it. I don't know whether 1% is fair or not—maybe it is—but I am pretty sure my opinion could change. As it stands, the legislature is powerless to change things without another statewide referendum. That's nuts. I could perhaps get behind something so permanent for a landmark civil rights issue, but not tax policy.

Second, ballot initiatives hamper compromise. And that's what government is, really: and endless series of compromises. You get your road, I get my park. Unfunded mandates are the worst: "You must spend $X every year. We don't know who's going to pay for it, whether it's necessary or produces results, whether it's fair, or whether there's a recession or other legislative priorities. Just spend it, every year, from now to eternity."

This isn't how we should govern. It's how we get disasters like Prop 13 (real estate) and Prop 8 (marriage equality).

Vote no.

SF Chronicle: California initiative process is out of control (Sept 7, 2018)

[1] Technically just the ad valorem (value-based) portion. We still have plenty of parcel taxes in Oakland of the form, "$300 for violence prevention", but they have to be a fixed amount that isn't tied to value.


Hospitals and video games

October 07, 2018

I had to go to the hospital today for some pre-travel vaccinations.

Every time I go to the hospital, or any medical facility, I think about usability. I'm fortunate enough not to spend much time in the hospital, so it's always a beginner experience when I do. Typically, I stumble around like an idiot trying to find the right floor, and that's assuming I'm even in the right building (not a given).

Whenever I get lost in a hospital, I think of how it's the opposite of a video game.

Nobody likes going to the hospital; we go there because we need something, often quite badly. But a hospital visit is a purely instrumental task: there is a because.

Video games have no because—in fact, many people wish they could play less of them, but they're so darn fun. Without any "purpose", they are the ultimate test of usability: is this so fun, I would waste time playing it? That's a really high bar.

A corollary might be that if you have something people need, your users/customers will put up with a lot of nonsense to get it.

But there are few things we need as badly as what's in a hospital, forcing most of us to do better that medical facilities, usability-wise.


Python: The new C

October 06, 2018

"If programming languages were countries, which country would each language represent?" My answer on Quora (Feb, 2015):

C: Ancient Rome. Widely studied, still the "lingua franca" of developers on whiteboards. Source of much modern culture, influenced everything, many great works of our trade found there/written in it. Primitive, somewhat barbaric worldview.

Just like Latin in its day, all serious universities teach C. It's the go-to for any new operating system, device driver, or language runtime (MRI, CPython, v8, CLR, JVM). Just like Latin, it's a "dead language" to some extent (though operating systems people might disagree), but its universality makes it the standard way algorithms are published and discussed.

But C is headed the way of assembly. Yes, everyone learned it, and no, nothing will take its place. But the real action today is in algorithms and data, not bit-slinging. Old-school "systems programming" is becoming more and more niche, with fewer and fewer people doing it.

Silicon Valley has taken a turn away from "programming" and toward true computer science. We've traded buffers, memory management, and bit-twiddling for data, algorithms, and math—and C for Python.

Python because, even though it's slow and synchronous, it has a great standard library, strong support for numerical computation, and reasonable automatic memory management. The innvation has shifted from bits and bytes to lane tracking, LSTM, and hidden Markov models.

I remember walking down 2nd Street in San Francisco, there was a whiteboard just south of Market St. that had been turned around, its back facing out the window. I walked by that office often and saw fragments of coding interviews. In 2010, I would've bet the language was a "brace language" (C, or one of its descendents).

Today, my money would be on Python, doubly so if it was a "data" role.

The future



October 05, 2018

The company recently remodeled the existing warehouse with an automated system created in partnership with Daifuku, a provider of material handling systems. Now that the system is running, the company revealed during a walkthrough of the new facility, Uniqlo has been able to cut staff at the warehouse by 90%. The warehouse can now also operate 24 hours a day.

- Uniqlo replaced 90% of staff at its newly automated warehouse with robots (Quartz)

People sleep; robots don't. That's a big change.

One immediate consequence: a "slower" robot that can work 24/7 may be faster than a person, who has to go home at night. I first thought about that when reading about SAM:

But I think work throughput is just the tip of the iceberg; a lot of life runs on human cycles, and I wonder whether any of the following will change, as robots become more widely used:

  • Will daily electricity use shift, if fewer buildings are occupied by people?
  • Will daily commute patterns change?
  • Will "construction season" (a big deal in the Midwest) last longer?
  • Will crop harvest patterns change?

Sometimes the second- and third-order effects of major changes are the most surprising.