Tuesday, May 26, 2015

Three-tiered system didn't help small breweries

The magazine for the Federal Reserve Bank of Richmond recently published an article about the history of breweries, which seems to have gotten one very important fact wrong. It repeatedly claims the three-tiered system, which prohibited breweries from selling beer directly to customers, actually helped small breweries. From the article Bottoms Up:
As a compromise reached during the repeal, a three-tier system was adopted: Brewers would now sell their beers to an independent wholesaler who would then sell the beer to independent retailers. The introduction of a middle-man to the market structure had several effects. First, it protected retailers from pressure from larger brewers to only sell certain products. Second, it provided small brewers with better access to the consumer, as the wholesaler would provide distribution equipment, marketing, and sales expertise — costs that could be barriers to entry for many small players. [...]

But the three-tiered system that helped small brewers in some respects wasn't enough to erase the advantage held by the firms that rode out Prohibition making syrups and malts. [...]

Additionally, the three-tiered system established after Prohibition is especially valuable to small brewers for getting their products to consumers through already-established distribution networks, according to an economic impact analysis conducted at the University of Delaware for the National Beer Wholesalers Association.
This simply doesn't fit with the historic record or what small breweries are currently say. A new paper shows that the more powerful the alcohol distributors are in a state, the fewer craft breweries exist. This is why one of the biggest legislative goals of craft breweries in many states is to relax the three-tiered system in ways to allow them to sell beer directly to consumers. There is a distinct positive correlation between states relaxing the three-tiered restrictions on direct sales for small breweries and the number of craft breweries that develop there.

This article even acknowledges that the breweries dramatically shrunk for decades under the three-tiered system and the trend only dramatically turned around when states began allowing small breweries a way around the three-tiered system.  From the article: "Then, in October 1978, President Jimmy Carter signed a bill that legalized home brewing and shortly after, individual states began legalizing brewpubs — restaurant- breweries that sell 25 percent or more of their housemade beer on premise. The craft brewing segment was born."

I don't know how you could conclude the three-tiered system has on net been "valuable"  to small breweries when looking at these trends and legal history.

Why this argument matters for legal marijuana.


A proper understanding of the impact of the three-tiered system is both important to the current state of the brewing industry but also extremely important to the emerging legal marijuana industry. As states legalize marijuana, they need to decide how to regulate it. To do that well, they should have a clear understanding of how alcohol regulations shaped that industry. What regulations are adopted could determine if the industry is dominated by only a few companies or by thousands of small producers.

*Image from Brewers Association

Friday, May 22, 2015

Fearing almost-intelligent machines, or the Blob

Concerns about the potential danger of true artificial intelligence have gotten significant attention lately thanks to statements from people like Elon Musk and Bill Gates. As I have said before, I believe humanity will handle this development with appropriate precautions because the Sci Fi community has done such a great job of warning us about the dangers of rogue A.I. There is, however, a somewhat related issue that I don’t think has reached the same level of public awareness: the danger of autonomous, almost-intelligent machines which lack self awareness.

With true A.I. I believe researchers already know to be cautious as they move forward. I believe they will try to hardwire them with morality and a respect for human life. At the very least, you should be able to communicate to an A.I. using logic and reason.

With stupid autonomous machines, the same precautions might not be taken. While we still don’t fully understand human style intelligences, we seem to understand swarm intelligence even less.

Take, for example, something like a termite mound. Individually, each termite is stupid and basically follows a few basic operations. Yet with thousands of termites all following these same operations, they create a level of complexity and destructiveness you could never predict from looking at one individual insect.

Imagine a company building an autonomous machine with what they think are a few harmless basic commands to take care of a simple job. They may fail to anticipate the high orders of complexity that will emerge once they have tens of thousands of identical machines working at the same time.

I believe it is highly improbable that humanity will be seriously harmed by our own robots. That said, I think a swarm of mindless recycling nanobots that unwittingly begin taking apart humans to refine or iron is a more likely risk than the emergence of a Skynet-style evil A.I. overlord.

True tragedies more often occur because of an unforeseen set of careless mistakes, not as part of a specific plan. If A.I. is ever created, it will be part of a very conscious and focused effort, but an out-of-control swarm is something that could just emerge from a set of seemingly unimportant oversights.

I think the upcoming remake of the classic movie, the Blob, could potentially be an interesting popular culture vehicle for talking about this concern. Instead of making the Blob an alien lifeform, they could make it an out-of-control stew of nanobots.

* Sergkorn at en.wikipedia [CC BY-SA 3.0 or GFDL], from Wikimedia Commons

Wednesday, May 20, 2015

The simple reason marijuana legalization is inevitable

Basic demographic changes, cultural trends, and the growing financial power of the legal marijuana industry are all pushing the country towards legalization. The only development with enough importance to have a reasonable chance of reversing this momentum would be if our limited experiments with marijuana legalization turned out to be policy failures -- or at least are publicly perceived of as failures.

That is why surveys like this new one from Public Policy Polling are so important. From PPP:
Also in 2012 Washingtonians voted to legalize marijuana usage by 12 points. Now voters in the state say they support marijuana being legal by 19 points, 56/37. 77% of voters say marijuana being legal has either had a positive impact on their life or no impact at all, with likewise only 22% claiming marijuana legalization has affected them negatively. 
Adult marijuana use been legal for over two years in Washington state and recreational marijuana stores have been open for almost a year. Direct experience with legalization hasn’t turned a majority of people in Washington against it, and they clearly don’t think it has been a policy failure. While it is still possible things might change with even more time for people to experience legalization first hand, it is currently looking very good for reform advocates.

With this new piece of data, it seems increasingly clear it is not a question of if marijuana will be legalized at the federal level, just a question of when and how.

Thursday, May 14, 2015

Disruption, incremental change and breaking the law

Google's Lexus RX 450h Self-Driving Car
 "Disrupt" is a popular buzzword in the tech world, but it lacks real meaning. New tech companies and products can typically fit into one of three categories:

Paradigm shifters - These are the truly rare developments that eventually create significant shifts in our whole society. One of the best early examples is the steam engine. By freeing people from the limitations of muscle or sail power, the steam engine changed the world.

A possible future paradigm shifter is self-driving cars. They have the potential to be more than just cars that are relaxing to travel in. If vehicles can drive themselves, it could eliminate the need for personal ownership of cars. Taxis would be far cheaper and more widely used, potentially freeing up huge amounts of urban space currently dedicated to parked vehicles. This could completely change the urban landscape. 

Incrementalists - Many like to think their ideas are revolutionary instead of evolutionary, but few ideas actually are. Most good ideas create only incremental improvements in society, but they are still important. Something that only slightly improves the lives of millions of people is still a significant net benefit to society and can be an extremely successful business.

One good tech example of this might be grubHub. Ordering food delivery has been a custom for decades; grubHub just made it marginally better. It has not completely changed how society interacts with food, but it has made a real improvement to an existing concept.

Rule breakers - A significant trend in the tech industry right now is "regulatory arbitrage." Effectively, a company pretends to have created real innovation, but their success mainly comes from violating the spirit of existing laws by claiming some obscure technical detail has created a new legal loophole. Sometimes these attempts to find a loophole fail. The biggest example of this is Aereo, which unsuccessfully tried to claim that by using millions of tiny antennas to retransmit broadcast television, it technically didn't function like a cable company -- and thus wasn’t subject to the same retransmission fees.

Sometimes these loopholes created by technology are found acceptable. Many of the most successful tech companies right now have combined real incremental improvement with regulatory arbitrage. Uber, for example, did improve car services with mobile payment and real-time tracking, but much of its success comes from exploiting legal loopholes exempting it from classification as a taxi company (despite behaving very similarly to one). This saves them from significant regulatory expenses. Similarly, Airbnb has made it easy for people to find places to rent, but early on it won a competitive advantage by their clients not paying hotel taxes and offering short term rents which in many places are technically illegal.

Companies that depend of exploiting loopholes, however, are always vulnerable to new court rules, counter lobbying, and the changing will of the government.

It would make for an interesting --but probably very difficult to conduct-- study to see if in the past several years the United States has seen more tech companies trying to exploit regulatory arbitrage than other first world countries given the incredible level of political gridlock in Congress. 

*By Driving_Google_Self-Driving_Car.jpg: Steve Jurvetsonderivative work: Mariordo [CC BY 2.0], via Wikimedia Commons

Wednesday, May 6, 2015

How long until the National Archives helps me get high?

There are two big milestones left for the marijuana reform movement in the United States. The first is when it gets begrudging acceptance under federal law. While the Obama administration has taken a relatively hands off approach to the issue, that is not a real long term solution; the industry will never be truly protected, respected, and legitimate until it is made legal under federal law.

The second big milestone is when marijuana use by adults reaches a level of true cultural acceptance -- a point at which marijuana prohibition is looked back on by basically everyone in the government as comically stupid. It will probably be impossible to know when this exact cultural tipping point is reached, but we know what it will look like.

Earlier this year, I was at the National Archives to see their special exhibit on the history of alcohol, and in the gift shop, they were selling things to help people get drunk. There were beer mugs, wine glasses, flasks and books with delicious cocktail recipes. Everything you would need for an alcohol-fueled party was prominently displayed in the front of the store.

This is what cultural acceptance looks like: When the idea that the government should try to stop adults from using marijuana is considered so misguided and antiquated that even government museum gift shops are selling souvenir bongs.

Monday, May 4, 2015

Sci Fi will save us from AI

A photo posted by Ex Machina (@exmachinamovie) on

Spoiler Alert: I’m going to give away the end of Ex Machina in this article. If you haven’t seen it already, you should since it is a very good movie. 

I deeply enjoyed the new movie Ex Machina. It did a great job dealing with the interesting issues about the future of artificial intelligence in a way that was accessible without sacrificing facts. Importantly, it was also just a very well made film with excellent acting and beautiful cinematography. I did have one problem with the end. 

It concludes with the creator of the first A.I. being murdered by his creation. It is a nice poetic end to the movie that's reminiscent of Frankenstein. The problem is it doesn’t make logical sense.

We are supposed to believe that the foremost expert on A.I. would have his only defense against one of his creations going rogue be an electronic lock. He didn’t use a manual deadbolt to lock her room. He didn’t tether her to a power source without batteries. He didn’t progam his creations with some form of Asimov’s three laws of robotics. He didn’t install a small bomb/kill switch in her that he could trigger with the press of a button or speaking a code word.

This is just silly because on this issue, the collective Science Fiction genre has succeeded. Sci Fi is best when is it preparing and warning us about things before they happen. When it comes to the potential danger of rogue A.I., we have been thoroughly warned. I suspect everyone in the country has at least seen a Terminator movie, a Matrix movie, or read one of the dozens of great Sci Fi books which have dealt with this concern.

Sci Fi writers have hammered into our collective consciousness that A.I. can be unpredictable and potentially extremely dangerous. I imagine everyone working on building the first true A.I. is going to include a manual kill switch and probably an extra dozen backup kill switches as a result.

That is why I’m relatively optimistic about A.I. -- not because I’m ignorant of the potential dangers, but because basically everyone in the tech world is aware of the potential dangers and will probably take the proper precautions.