Brexit: Support Theresa May. At Least for Now.

I argued earlier that a major flaw with the Brexit referendum was lack of any positive program in case of leave wins. The current political spectacle in London is an attempt to define such a program, and the withdrawal deal that Theresa May has negotiated is one related proposal. It satisfies no-one. But it is the only realistic compromise we are now left with.

The central problem with Brexit

The central problem with Brexit now, 2.5 years after the referendum, is still the same: Britons cannot agree what "leave" means. The country is fairly equally split between leavers and remainers, and unfortunately these two opposite preferences are hard to reconciliate. The options in the middle, along the lines of the "Norway option", satisfy only a few.

I think the best way forward in summer 2016 had been simply to take time off. Theresa May could have started her prime ministership by asking: Well, voters. You want out? So let's see what kind of options we have here, and what do you think about these. And waited with triggering the article 50 until she had an answer to these question. Unfortunately, this was not the case. There were no attempts to start a serious discussion about what Brexit may mean, and what kind of compromises are possible.

I do not see such attempts now either. The prime minister's repeated claims that there will be no second referendum sounds like "I don't care about which Brexit you want." Moreover, accusing those who are worried with no-deal outcome for torpedoing the negotiators and being part of the "project fear" is neither an invitation to a serious dialogue. Instead of trying to build a consensus, at least a somewhat consensus, the leaders are pushing to implement their own favorite flavor of Brexit. This has not worked well, but mainly because there is no consensus in the Conservative party either, and not because of concerns regarding the "will of the people".

It is not just that Britons cannot agree on what does leave mean. A large number of people do not want to leave EU at all. Anti-Brexit march at Conservatives' conference, Oct 1st, 2017.

Take time off now

But I don't think the second referendum is a good idea. It would suffer from some of the same flaws as the first one. While we now can offer a positive program, for instance "no deal", "May's deal", and "remain", the majority of votes will probably go to both of the extreme options, and that will give a closely split vote again. Hence there will again be no clear mandate about how to proceed. But the political battle over the exact wording of the question, the pre-referendum campaign, and post-referendum interpretation will deepen the existing cleavages, increase resentment and undermine support for the democratic institutions in Britain.

I think what UK needs now instead is just time off. It cannot have time off as EU member any more, at least not without a rushy U-turn and without badly hurting the feelings of the large group of leave-minded people. I think we should have a second referendum. But not now. Let's the smoke clear and contours of the post-Brexit world appear. Let also one of the options garner clearly more support than the alternatives.

But so far, let's take Theresa May's deal. No one is happy with it, I know. I am not happy either. But it is a realistic deal, and it allows UK to stay in the EU-s "waiting room" while it attempts to understand if it wants to enter again or leave for good. Call this "transition period". But the transition is not about the trade negotiations, but about UK trying to transition into a country that can somewhat agree on it's future.

What is wrong with referendums vol 2: Brexit

In a previous post I discussed some of the problems with referendums. Brexit referendum had it's share of issues. I'll discuss here two of these.

Restricted categories of voters

First, it violated the principle that everyone who has large stakes in the process can have a say. Two groups that are most affected by British EU membership and it's potential termination are the EU citizens living in UK, and UK citizens living on the continent. Unfortunately, these two groups had limited voting opportunites. British expats can only vote up to 15 years after leaving the UK, and Non-British EU citizens (except Irish, Maltese, and Cypriots) were excluded by a political decision. Maybe exaggerating a bit, but one may say that only those who weren't using the EU provided free movement were deciding it's fate.

In a broader sense it is a common situation: migrants usually cannot vote (despite paying taxes) in their new host countries, and often find it hard or impossible to vote in their countries of origin as well, so migration-related policies are decided by people who are not moving. This is partly because of design, insiders are keen to keep their exclusive rights, but partly due to insiders' inherently better knowledge and better networks.

No positive question

Second, the referendum asked a "negative" question—there was little indication what will happen in case "leave" wins. If "stay" had won, the answer had been simple—more of the same. But leavers had no positive program, and one can argue that leave voters may have voted for very different and incompatible ideas.

One of the central issue with the Brexit referendum was lack of any positive program. Two years later, the UK government is still debating what does "Leave the European Union" mean.

If one disregards the slight margin, one can still argue that leaving the EU was "will of the people". However, one cannot extend this claim to any particular type of Brexit, or to any particular reason, such as immigration. These claims are either based on polls or some other kind of interpretation. But polls and interpretations are just that—polls and interpretation—and not the "will of the people".

It looks like Britons have currently at least two incompatible "wills". Because of the slight margin and blurred options, the referendum gave a mandate not for leaving the EU, but to bring the UK and EU relationship back to the drawing board. This may mean leaving the EU but that should be a different decision, taken years later, and including a positive program for what will happen in that case.

I published a post about R and openMP in http://www.parallelr.com/:
http://www.parallelr.com/r-and-openmp-boosting-compiled-code-on-multi-core-cpu-s/

What is Wrong with Referendums? A Reply to Geert Wilders

Recently (2015-11-19) Geert Wilders published an article in New York Times arguing that Europe should hold national referendums on immigration policy. But referendums are not necessarily fair, nor are they the ultimate form of democracy. Referendums have two problems: first, only insiders can vote, even on questions that largely concerns outsiders. Second, everyone has a single vote of equal weight, even in case where the importance of the outcome differs widely.

Who Can Vote?

Current "universal" voting is limited to "insiders", those who are either confined to a certain territory, or possess a certain legal status. In case of national referendums, these are typically citizens of the country; in local referendums these may be residents of a city. Outsiders cannot participate, even when voting over issues that have major consequences for them. This violates a central pillar of democracy, the ability to influence decisions that affect you.

A referendum over immigration policies is just such an example. Unless potential immigrants also participate, we cannot talk about universal democratic decision. Similar issues arise when neighborhood residents vote for restrictive zoning laws. The potential "immigrants" to the neighborhood who are adversely affected cannot participate because they are not residents. The Greek referendum of 2015 shared similar traits: only Greek citizens were voting over a decision that involved most of the EU.

Greek 2015 referendum ballot. The referendum had several problems, only Greek citizens voting over a pan-European decision was perhaps it's most serious flaw.

How Big Is the Vote?

Another problem with referendums is the "size" of vote. Traditionally, everyone has a single vote. This is true even if different people have very different stakes in the question. In this way the majority can always outvote a minority. For instance, in many places the majority may outlaw homosexuality. The Swiss minaret law is an outcome of a referendum that was flawed in this way. I don't know how many Muslims and non-Muslims actually care about minarets, but I can imagine it is a far less important issue for non-Muslims. But they could easily muster a majority. As an extreme example, imagine that Hitler put Nuremberg Laws, the Nazi laws that robbed Jews of their rights, on a national referendum. Would we consider the outcome democratic? Analogously, potential migrants may outvote the whole EU if given an equal weight.

Solutions

Unfortunately, these two problems cannot easily be corrected. Although it is not fair to exclude outsiders from decision–making, and give equal say for those with unequal stakes, we cannot calculate the fair vote size. Referendum is justified when the electorate has roughly equal stakes and outsiders are little affected. If this is not the case, one should not call for a referendum.

Instead, we may strive toward inclusive representative bodies that also involve outsiders. In case of the current refugee crisis it should include representatives from both EU, and refugee origin and transition countries. Such conferences, potentially meeting regularly, would ensure that insiders have more rights but outsiders will also have a say.

Britain in EU: More Powers to National Parliaments?

More power to national parliaments is a democracy–enhancing move only if the national decision-making occurs in areas that have little impact on the others. If the opposite is true, it diminishes the control we have over the important decisions.

*

David Cameroon has announced his requests to the EU. One of them is a larger role for national parliaments in shaping EU regulations. Although national parliaments may be the most democratic institutions we have, the areas where such a "repatriation of powers" occurs must be chosen carefully. Otherwise it may weaken the democracy instead. There are two main reasons.

Just a common market—or an "ever closer union"?

Interdependence

First, and most importantly, only some decisions can be made by national parliaments. Many decisions can not, such as international treaties, trade agreements and international law. Almost by definition this type of legislation is not made by individual states. Analogously, parliaments do not vote on what the other countries do, be it introduction of incompatible regulations, or conducting irresponsible fiscal policy in the hope that we will bail them out later. Neither have we much room to vote about technical standards, such as Apple's new operating system or airport security procedures.

A number of major problems EU currently faces cannot be solved at national level. For instance, the refugee inflow must either be received, or stopped, at the EU border. The current country-based decision making essentially allows Hungarian parliament to decide over number of refugees in Serbia and Croatia, and, if German "national parliament" decides to close it's borders, it has severe impact on Greece and Italy. Analogously, the "single market", one of the pillars of EU, is a form of extended free-trade agreement, a large number of common standards for product labeling, food safety, and labor treatment. By definition, agreement is something we cannot do alone.

More powerful national parliaments in this type of decisions will weaken EU without strengthening the member–states. We will notice more small inconveniences, such that you cannot use certain mobile services in another EU country, or your business has to hassle with incompatible regulations across the border. Negotiations among 28 sovereign states are far more slow and costly than central decisionmaking, and there is little incentive to overcome even small special interests in the name of a common good. National politicians are elected to stand for "national interests", and all 28 national interests are seldom aligned with the common one. The underlying problem is interdependency, our decisions may influence others even if the others do not belong to our "nation". Such decisions should be done by inclusive higher level bodies, such as European Parliament.

Enfranchisement of EU Migrants

The other, and currently less important reason, is enfranchisement of EU migrants. This is a large group of people who cannot vote in national elections but is subject to the corresponding national legislation. This is simply not fair. I think the first-best solution would be to give EU migrants voting rights rather soon after moving to another country (say, in 2 years). But until it happens, decision–making by EU institutions offers this group more say about their lives.

*

Finally, even EU is too small for many important decisions. Many contemporary problems, such as global trade agreements, climate change, or the puzzle of Middle–East, cannot be solved at EU level either. We need global governance more than ever before. How to achieve this in a democratic way is one of the big challenges of our time.

Improving the Pedestrian Experience: The Cheap Way

Traffic improvements may sometimes be cheap. Below I show how to enhance the walkways around a University of Washington (UW) building, Schmitz hall, in Seattle at University Way and 41th street.

A bit of background. University Way is a lively shopping street one block west of the UW campus. It is a major pedestrian destination and also an important connection street. UW campus begins one block east of it, on the other side of 15th avenue. At the end of Campus Parkway (40th street) there is a much used walkbridge over 15th. Schmitz hall is located just between the walkbridge and University Way and hence a major walking corridor embraces it from both sides.

Let's start from north, approaching the campus from University Way (see the picture). We don't want to follow the sidewalk straight down (south)—then we have to climb up later again to reach the walkbridge. We take instead the level walkway around the hall, either by turning left, or south. However, as you can see, the most direct route to the walkway is blocked by a small decorative wall. This is the first cheap improvement: just remove the wall.

NW corner of Schmitz Hall at University Way (right) and 41th street. The walkbridge is at the SE corner of the building, behind it at left.

Compared to typical Seattle sidewalks, the walkway is surprisingly spacious. Unfortunately, a half of it is blocked by short stairs, leading to the sidewalk at about 1m above it (see the picture below). This makes a large swath of the walkway unsuitable as a connector route. I don't know if it is used for anything else. Note that even the narrow gap between stairs and a concrete pillar is blocked by a trashcan.

The walkway on the north side of Schmitz Hall. The wider left-hand side is mostly blocked by stairs (middle-left on the picture).

But the next picture indicates just how easy it is to recover almost half of its walkable width: a) move the bin; and b) remove the lower part of the concrete wall that lines the stairs. Metal handrails are enough for safety. You may even shorten the handrails a bit at the lower end. This is easy and cheap.

A trashcan is placed in the narrow passage between stairs and a concrete pillar.

The next obstacle is a bike rack. Well, not the rack itself but the bikes that extend far out of the rack. Fortunately the solution is easy: turn the rack 90 degrees and move it over to the left side.

The bike rack is placed gently between the columns. Unfortunately, the bikes occupy most of the walkway.

After passing the rack we reach the bridge with no further obstacles. However, there is another issue that can easily be solved. A bus stop is located at the 15th avenue, just at the end of the footbridge. Unfortunately, as the street is about 1m lower than the walkway at the end of the footbridge, one has to walk around the waiting shelter ("out of the picture" on right-hand side). This adds about 15 seconds to the walking time compared to the direct route.

Those arriving from campus over the walkbridge (visible in the background) must walk around the shelter to get to the bus. One should construct stairs at the direct line between the bridge and where the bus stops—in the center of the picture. The bus stands on 15th avenue, a wall of Schmitz Hall is visible in the upper right corner.

15 seconds feels rather long when the bus is just coming, and as a handy shortcut, many jump the wall instead. Indeed, this is a place where one can easily add small stairs.

Landscaping beneath the wall indicates that many passengers do not take the trip around the shed but just jump.

*

We need more attention to small details when desiging transportation infrastructure. The examples above indicate that improvement can sometimes be achieved with very little costs, or even at no cost if such considerations are taken into account in the original design. We need more awareness and understanding of pedestrian mobility at least as much as money. In this case it is about directness and space: we don't want to go around if a direct route is just here, and we don't like narrow passages.

*

AD ENDUM: I moved the bike rack out of way.

Bike rack moved out of way

Gas Tax May Be a Simple and Good Enough Road Usage Charge

US road maintenance is underfunded. Unfortunately, current politics does not offer any sustainable funding ideas either. Instead, we hear about various one time patches like playing with custom tariffs or selling strategic oil reserves.

A good funding mechanism is based on the following principles:

1. It creates a steady income.
2. The Income grows and falls according to the road maintenance needs. This means it is related to the total driving mileage.
3. The funding burden is closely associated with individual road usage, in particular the damage one causes to the road structures when driving.
4. The mechanism is largely inflation-proof and does not require frequent political interference.

Freeways permit fast uninterrupted travel through dense urban environments. But they are not cheap. I-110 in LA.

By Adrian104 [Public domain], via Wikimedia Commons

In recent years we have seen an increased interest for "road usage charge", a user fee that depends on the actual driving mileage. Indeed, modern technology (currently tested in Oregon) allows to determine the exact mileage on different types of roads, and to send the driver the bill afterwards. Here I argue that we do not have to wait for the a technology to mature, as old-fashioned gas tax may serve as a good substitute for a smarter driving distance fee. I solely focus on road maintenance funding, and ignore congestion, pollution, and climate issues. Note that from this perspective there is no difference between fossil fuel, biofuel, and electric cars.

*

Two arguments strongly favor gas tax over alternative funding mechanisms: it's simplicity and it's focus on road users. The main objections are related to it's impact on economy, and the fact that it is less than perfect measure of actual road usage.

• Gas tax is simple to introduce, collect and pay. First, the direct payments are done by big oil market operators who can easily handle a rather minor additional administrative burden. Second, most governments already collect it, hence the additional administrative burden would be even smaller. Third, such "pay-at-pump" scheme is the simplest imaginable tax for motorists. You pay for your gas and that's it.

  This contrasts to the proposed distance charge which is to be payed individually by millions of drivers. Hence the aggregated administrative burden for both private actors and public administrators is most likely higher. The distance tax must also be payed monthly or yearly, based on the actual driving, in a similar fashion as we currently do with the electricity bills. This also means the drivers have to keep some funds available for the tax payments later.

• Gas tax is payed according to road usage—the more you drive, the more you pay. Unfortunately, this correspondence is less than perfect. Cars come in different fuel economy and size and stress the roads differently. This is potentially the main objection against funding roads solely by gas tax.

  

  Heavy transport stresses the roads substantially more than small cars. Although trucks burn more fuel than cars, the gas tax may not compensate for the additional stress on roads. Here weight-dependent usage charge may have an advantage.

  By bilbobagweed (formby) [CC BY 2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

  But save the road usage charge, gas consumption is still far closer indicator of individual "road consumption" than any other alternative, such as income or sales tax. If gas tax is too crude a measure, how on earth can sales tax be a better one? But sales tax is widely used for funding transportation projects.

• Gas tax is often claimed to be a burden to the economy. But the picture is more complex. Sure, taxes hurt, but I don't see why should gas tax hurt more than the other taxes, in particular business and payroll taxes. If we introduce it in revenue neutral way, i.e. we lower the other taxes by exactly the same amount as we rise the gas tax, it amounts to redistributing the tax burden from the rest of the economy to large gas users. It is not immediately clear what are the economic implications. Transport intensive sectors will probably lose (but it also depends on what happens with road quality and congestion) while "human-intensive" sectors win from lower income tax. This includes technology companies that rely on a large well-payed workforce. I do not see the effect being much different from that of a better targeted road usage fee.

• Inflation diminishes the value of both gas tax and road usage charge in a similar way. Unless inflation-indexed, regular political decisions are needed to rise these accordingly. The usage charge possesses a clear advantage here as it does not depend on the vehicles' fuel economy. Gas tax must be adjusted both for inflation and fuel economy, usage charge only for inflation.

• Finally, many people may dislike the idea of government knowing exactly where and how much we drive. And the corresponding technology itself is not safe either—the ways to screw such meters will probably advance as well.

*

We do not have to wait until a better technology solves the road funding problems. It may never arrive. Meanwhile, gas tax is a simple and good enough road usage fee. The problem is in politics, not in technology.