2018-11-01

Solving Australia's House Price Crisis

The Problem

I used to believe that one of the most solid economic laws that can ever be argued for is the notion that "what goes up must come down". In other words, asset price bubbles (such as property or sharemarket over-investment) will naturally fall back to realistic levels. Tulip mania, the 1929 Stock Market crash... all the evidence is there.

But no longer. Asset price growth in property and shares has continued. Even the correction inherent in 2008 Global Financial Crisis has been exceeded in some areas. I am referring especially in this case to Australia's property price bubble.

Current Monetary Policy, which has evened out the business cycle, is probably to blame here. Controlling levels of inflation by adjusting interest rates has been one of the success stories of the world's post-1980 economy. But modern measures of inflation don't take asset price growth into account. A low inflation economy with a growing asset price bubble has been the norm in many western economies for decades.

One Solution: Higher Interest Rates?

One solution would be to factor in asset price bubbles into inflation measurements. As an indicator, this is fine. But what should be the central bank's response to this changed rate of inflation? Would it be wise to increase interest rates to push down a growing asset price bubble when the rest of the economy probably won't handle such a change? After all, if, say, the Reserve Bank of Australia (RBA) increases interest rates to deal with the growing property bubble, would the effects on the non-property sector of the economy result in a bad recession? Would the cure be worse than the disease?

Of course the real problem here is that interest rates have a very broad effect on the economy. This is inherent to current monetary policy and it can't be avoided. This doesn't mean that adjusting interest rates are somehow a bad thing - there are obviously times when a broad adjustment is going to be required.

But in the case of asset price bubbles, especially in the case of Australia's property market, a different approach is needed. One in which specific monetary goals in that particular market need to have external adjustment. A micro-prudential policy, probably with a dash of micro-monetary policy as well.

A Better Solution: Keep Property Prices Stable

In the case of Australia's overvalued (and now deflating) property market, it is important to neither keep prices rising nor to let prices fall. Property should retain its value over the long term, adjusted for inflation.

For this, an inflation-adjusted property price index should be created, with regular monthly updates. With the index starting at 100, the goal should be a long term price index that has peaks and troughs, but remains at 100 on average.

The advantage of stable property prices is threefold.

Firstly, it means that those who invest in housing will be investing in something that retains its value. All forms of investment exist because people with money wish to gain more. If property prices retain their real value over the long term, people will feel safe investing in them. Obviously such investment needs to be compared to other forms of investment, and the risk/rewards that such investments have, in order for the market to respond appropriately.

Secondly, it creates a disincentive for speculation. Any form of asset price growth leads to speculators who are not so much concerned with actual assets, but whether they can profit from the process of buying low and selling high. Speculation has its place in an economy, but asset price growth in shares or property turns investors into speculators. This results in profits derived from the process rather than the utility of the asset being invested in, and creates a parasitic sort of investment class. This has been one of the great problems in the world's post-1980 economy, with huge asset price bubbles in both shares and property coexisting with lower levels of GDP growth. But if an asset is neither growing nor shrinking in real value over the long term, there is less incentive for speculation.

Thirdly, assuming wages keep rising, it makes for more affordable property prices over the long term. If an economy has rising wages but property prices remain stable, then, by definition, property prices become more affordable. My belief is that property prices are too high, and there are plenty of stories and statistics out there which show how difficult it is for younger people to afford to buy their own accommodation. A long term increase in housing affordability is the best solution here.

So how would this be achieved?

The real property price index would, like the inflation index, be used by a central bank to determine whether to remove or inject liquidity into that specific market. Currently, the adjustment of interest rates is the only current solution, and as I have pointed out above, its usefulness is in its broad effect, not in its narrow effect, making its use problematic in the case of stabilising property prices.

The solution would require a central bank to use two policy tools to keep prices stable. Micro-prudential policy tools would be used to stabilse the peaks and troughs of a typical business cycle, while micro-monetary tools would be used during a serious drop in prices.

Micro-prudential tools

If the index is showing a substantial change in house prices, then the Central Bank would adjust prudential rules that govern mortgages, by increasing or decreasing the minimum deposit required for mortgage holders. This would be based on lending laws that would apply across the entire property market, ensuring that people applying for a mortgage would have to increase or decrease the initial deposit.

In a market of increasing property prices, these micro-prudential rules would have the effect of denying liquidity to the property market, resulting in less money available to invest, and thus cause a drop in prices.

The same micro-prudential policy can be used when prices begin to fall. By decreasing deposit ratios, more liquidity from investors would enter into the market, causing prices to rise.

Micro-monetary tools

In the case of a substantial drop in prices, however, direct monetary policy would be required. Even if micro-prudential deposit rates were dropped to zero, economic conditions might still be so bad that prices keep dropping. When this occurs, the central bank could invest directly in the market itself. This would involve the central bank creating money by fiat, and then using it to purchase property. This is similar to Ben Bernanke's "money by helicopter" theory, except that instead of giving free money out to everyone, fiat money is used by a central bank to directly enter a market. Once prices have begun to stabilise at the 100 index level, the central bank can then begin gradually selling off this property, with any money it gains from the process being "de-fiated" into nonexistence.

Of course such a targeted policy would impact the wider economy, and inflation rates will be impacted by the specific policy tools that the central bank would use to stabilise prices. But these broader effects would best be served by broader policy tools - ie, interest rates.

Summary

To summarise, the current property price bubble in Australia can be solved through the following means:

1. Set up an index which measures property prices adjusted for inflation and update it monthly. This would require work on behalf of the Australian Bureau of Statistics and funds to create it.

2. Grant the Reserve Bank of Australia the power to determine mortgage deposit rates, requiring all registered mortgage lenders to set a minimum rate of deposit. If the rate is set at, say, 20%, this would mean that someone wanting a $1 million mortgage would have to have saved a $200k deposit.

3. Grant the RBA the power to change mortgage deposit rates.

4. Give the RBA the directive that mortgage deposit rates should be adjusted in order to keep real property prices stable over the long term. Specifically, this would be an index number of 100 averaged out over the long term.

5. Grant the RBA the power to purchase property with money created by fiat as another way to keep real property prices stable.

6. Property so purchased by the RBA will be maintained and managed by a separate government body until such time as the RBA sells the property.

7. Property directly purchased by the RBA will be sold once prices have stabilised.

8. Money generated by the RBA's selling of properties is "de-fiated" into nonexistence, and is not part of general government revenue.

2018-09-09

The danger of Arctic Methane

Over the past few years I have been examining the issue of Arctic Methane emissions as part of the effects of global warming. It's an issue that is growing in importance, but also an issue that is deeply misunderstood and misrepresented by the media, and which has attracted far too many conspiracy theorist "climate doomers". As a result of this mishmash of falsehoods and conspiracies, the issue of Arctic Methane does not have a very good reputation among mainstream climate scientists, and is sometimes used by climate denialists as evidence of crazy unscientific behaviour by advocates of global warming.

Of course I am not a scientist, which means that the only thing you, the reader, can truly rely upon is to check my statements and facts with known science. I'm reasonably confident that anyone who does this will find my presentation here to be correct.

The situation, in a nutshell, is as follows:

As the Arctic regions warm up, undersea and underground methane reservoirs will increasingly begin to out-gas into the atmosphere. Methane is a very potent greenhouse gas, and increases in atmospheric methane will significantly speed up global warming and its effects.

A further explanation of the situation:

In the Arctic regions of Northern Canada and Siberia, there exists a layer of permafrost which acts as a "lid" or a "seal" over vast reservoirs of methane gas. This is especially the case in an area known as the Eastern Siberian Arctic Shelf (ESAS), an underwater shelf in the Arctic Sea to the North of Eastern Siberia. This is an area of reasonably shallow sea, with a mean depth of 50 metres. Below the bottom of this shallow sea is a layer of "subsea permafrost" which has remained permanently frozen for many thousands or even hundreds of thousands of years. Beneath this layer of subsea permafrost is what is known as a "sedimentary basin" - a geological feature which is essentially made up of sediment depositions over a very long time (most likely to be from the many rivers that flow out of Eastern Siberia into the Arctic Ocean). Along with river sediments, the basin was also filled up with the biomass of plants and animals from ages past. Over time, this biomass was subjected to heat and pressure, and turned into methane.

The situation now is that heat from global warming is now beginning to melt the ice cap. Data from the past few decades has shown that the Arctic ice cap is quickly melting, and there is a reasonable expectation that the ice cap will completely melt during a northern summer some time in the next 10-15 years. Phase Transition - the moving between solid and liquid states of matter - is a process that actually takes a lot of energy to accomplish. Once the ice cap has melted and phase transition is no longer an issue, the temperature of the Arctic ocean will begin to steadily increase. Moreover, a greater mixture of the water column will occur due to the effects of wind and waves - presently this process is prevented over much the north pole during winter by the presence of sea ice, which stratifies water temperature. With a greater mixing of warmer water with water from the bottom, a heat pulse will begin to melt the subsea permafrost. As this subsea permafrost melts, cracks called gas migration pathways will open up, leading to a release of the methane trapped under the ice seal into the water. With only 50 metres of sea level, the methane will quickly exit the water and enter the atmosphere, where it will spread over the globe.

It is estimated that there are 5 gigatonnes of methane currently in our atmosphere. It is also estimated that the ESAS contains 100s to 1000s of gigatonnes of Methane beneath the permafrost seal. The chances are that 50 gigatonnes of methane could be released into the atmosphere within the next few decades, at most 100 years. A tenfold increase in atmospheric methane would lead to a major acceleration in warming.

The best explanation I have read about this issue comes from the Russian scientist Natalia Shakhova, who was interviewed about this subject and explains it here. Shakhova is probably the best person to read and listen to about the subject because she and the others who work with her regularly go into the Arctic to measure undersea methane levels and publish papers about the subject. Shakhova is the source for the 50 gigatonne output.

Unfortunately, Shakhova's work has been misrepresented and sensationalised by the media, as well as by the "doomers" and conspiracy theorists. The impression has been that the outgassing of methane is something that would happen suddenly, as though Shakhova's 50 gigatonne prediction might happen tomorrow. Vivid imaginations (like mine) could imagine a massive cloud of methane suddenly bursting out of the Arctic ocean, rising to cover the surface of the earth and quickly turning our world into Venus-like hellhole in which no form of life can survive. People who follow this apocalyptic vision then begin to accuse the IPCC and NOAA of holding back vital information, and that the statistics regularly released on atmospheric methane and global temperatures have been doctored to prevent global panic. Chemtrails and the Illuminti probably get mentioned as well.

It's this group, the doomers, that have probably done more to damage this theory than anything else. When I began to study this issue, I contacted climatologist Gavin Schmidt, who kindly responded to my questions promptly. His view is that the methane danger is overblown and its proponents are peddling bad science to scare people. Notably he didn't include Shakhova in this response. When I began to review his comments later on, I realised that he wasn't disagreeing with Shakhova's study, but with the bad science of the doomers. I also realised that he probably did not know the actual situation, which I described in the first few paragraphs.

One thing which has coloured the argument is the use of the word "clathrate". In the initial popular understanding of this subject, the idea was that methane clathrates - a combination of water and methane that forms a solid ice like substance - were the threat. Schmidt rightly pointed out to me that the majority of the world's undersea methane clathrate would not be subjected to such high temperatures for a very long time, because the majority of it was much deeper in the ocean, and the heat pulse from warmer water would take a very long time to reach it.

Nevertheless, the 50 gigatonne figure from Shakhova is focused solely upon the unique conditions of the ESAS, which is that a store of methane lies beneath a layer of melting subsea permafrost. How this methane is stored - whether it is a hydrate form or as a liquid under pressure - isn't the issue. Climate scientists like Schmidt are probably more concerned about sensationalist headlines and whether the "Clathrate Gun Hypothesis" is being misunderstood. Of course they should be concerned that bad science is exposed, but in this case I think they haven't done the actual issue - that presented by Shakhova - justice.

So what would the impact of a 50 gigatonne methane release be? A 2013 study, "Vast costs of Arctic change", written by scientist Peter Wadhams and economists Chris Hope and Gail Whiteman, estimated that the total cost to the world economy would be $60 Trillion (2012 figures), which is essentially the size of the entire world economy. In terms of the heating effects on the atmosphere, Hope stated "The methane release would bring forward the date at which the global mean temperature rise exceeds 2 degrees C by between 15 and 35 years,".

So when will this occur? The answer is that it has probably already begun, but the amount of methane that is being released is too small to make any real difference. Certainly the amount of atmospheric methane has been steadily rising for many years now, but this is due to a large amount of factors (cow burps, fracking, rice farming) rather than the result of Arctic methane.

What we can predict is that the process will begin to speed up once the North polar ice cap completely melts during the northern summer. The polar ice cap, which floats on top of the Arctic ocean, has been steadily decreasing in mass for many years now, and, as I stated above, is likely to fully melt some time during the next 10-15 years. Some scientists have named this predicted occurrence "A blue water event", meaning that the Arctic ocean is completely free of surface ice. Of course ice will grow back during the winter months, but the net effect will be less ice, less albedo and more heat storage in the liquid water, and a water column mixing warm water into the shallow bottom of the ESAS, all resulting in a weakening of the subsea permafrost and an increasing amount of methane finding its way into the atmosphere.

And of course the effect of this outgassing will not be a sudden event - it will occur over many years, getting worse as time goes by. We'll be able to detect it via methane detection equipment in both the Arctic (Barrow) and in the Equator (Mauna Loa), both of which are in place and have been recording atmospheric methane levels for many decades. There are also satellites currently measuring atmospheric methane levels.

But what are the variables? it's one thing for me, a non scientist, to present this situation. It's another to be objective and judicious about it.

The first variable has to be the size and strength of the sub-sea permafrost. The permafrost "lid" that Shakhova describes as being under the ESAS cannot be uniformly thick, but must be thicker in some places and thinner in others. Obviously the thicker permafrost will take longer to melt. Moreover, if the thickness of the permafrost is determined to be larger than is currently measured, then its melting will take longer, and the release of methane will be more constrained. Obviously more study needs to be done in this area.

The second variable has to be the size and location of the methane. Like oil reservoirs, methane exists in permeable rock. This means that the methane itself is also unlikely to be a uniform shape and thickness under the permafrost. Recent land based methane outbursts in Siberia show that the underground gas seems to be centred in pockets. Obviously the surrounding geology, affected by the thickness of the permafrost lid, is likely to affect the time and exitent of outbursts.

The third variable is the speed at which the permafrost melts. As I have pointed out above, warmer waters and a lack of sea ice will inevitably lead to a melting of the subsea permafrost. But at what speed will this pulse of heat travel at? And what of the experience of phase transition that I have also mentioned, namely that it takes a lot of energy to turn ice into liquid water.

Conclusion

At present, the danger of undersea methane in the ESAS is considerable. The effects that it could possibly have upon our planet are dire. More research into the ESAS is desperately needed. Geoengineering to specifically cool the Arctic is also needed.

2018-09-06

Donald Trump and The 25th Amendment


Any attempt to remove Donald Trump from the presidency using the 25th Amendment is likely to fail

Impeachment is actually more likely to work, as I will shortly point out.

The 25th Amendment was written into the US constitution as a way to determine who runs the government if the president is incapacitated. This would mean that the president is so badly injured or so badly ill that he is incapable of functioning in his position. The president's cabinet, or a committee set up by congress, would then send an official letter to congress informing them of the president's incapacity. The Vice President would then be appointed into the executive role until such time as the president recovers. Once he has recovered, he would then begin working as president while the Vice President returns to his position.

Section 4 of the 25th Amendment, however, cover the possibility of a president potentially going mad or being incompetent. While these terms are not used in the amendment, it is clear that its writers assumed that a time might come when the president is unwilling to give up his role, temporary or otherwise, in the face of a 25th Amendment removal.

When this occurs you then have an impasse. If the cabinet remains committed to remove the president but the president remains committed to staying there, then the decision is referred to congress.

And in order for congress to remove the president, a two-thirds majority in both houses is needed to succeed. This is why impeachment is easier (it needs a majority in the house and two-thirds in the senate).

So if Trump's cabinet does theoretically decide that the president should be removed, the following would occur.

  1. The Cabinet would write an official letter to congress stating that it is in their considered view that the president is incapable of functioning in his role. As a result, they are appointing the Vice President, Mike Pence, into the president's position. 
  2.  Once this has been done, Trump would then write a letter to congress. In this letter he would say that he is fine, is the most intelligent person on the planet, and quite capable of functioning as president. Once this has been done, he would then go back to being president, and Mike Pence would return to being the VP. 
  3.  After this, the cabinet would send another letter to congress, saying that Trump is not capable of functioning as president. Once this has been done, the power to make a decision is given to congress. 
  4.  Congress now has to decide whether to remove the president. Two-thirds of both houses of congress are required to make this occur. As this is going on, Trump supporters are protesting outside congress, warning the Republicans in congress to not remove the president or there will be hell to pay. Some Republicans vote to remove him, but the magic two-thirds line is not reached because there are enough Trump supporters in congress to prevent the removal. 
  5.  Trump remains president and fires his cabinet. Mike Pence resigns. Trump appoints Jared Kushner as VP.
The only way for Trump to be removed from office using the 25th Amendment is if congress makes that decision. And that decision is harder to make than a decision to impeach.

2018-09-02

Citizen Politicians: Selecting Politicians Randomly

I just watched a TED video by Brett Hennig titled "What if we replaced politicians with randomly selected people". It's a great idea, and one which I proposed way back when I began this blog in 2005. But having read my original post, I think it is time to update some of the changes in thinking I've had since then.

One of Hennig's great actions during this TED talk is to ask two questions to the audience. The first was "Who thinks living in a Democracy is a good thing?" Answer? Pretty much everyone in the auditorium. The second was "Who thinks our democracies are running well?". Answer? Pretty much no one.

Since my original post in 2005, I am convinced more than ever that modern democratic governments are fatally flawed, mainly because the people we elect to make decisions for us are manifestly unsuitable. And this applies across all wings of politics. We expect our politicians to work co-cooperatively in parliament or congress when making decisions, but the way we select our politicians requires competition and advertising. Moreover, we cannot expect everyone in society to have the same interest or understanding of politics, which means that people will often vote for a party based on tribalism (like a sports team you support through thick and thin) or else they will vote according to what a lot of their friends, family and/or work colleagues will say. Sometimes this is justified, especially if a political party or its representatives have made bad decisions. But sometimes it is not justified, especially if a competent political party has been misrepresented using modern advertising.

The latter situation - a political party being misrepresented by an effective advertising campaign - was something I fell for when I was young. In 1996 I voted for John Howard and the Liberal Party, mainly based upon their message that the Australian Labor Party had led Australia into a dangerous fiscal crisis. Many years later, when crunching numbers and statistics on Australian government debt levels, I came to the inescapable conclusion that the Liberal party message back in 1996 was a falsehood.

Advertising and propaganda is increasingly sophisticated. Lies are not punished. Instead the two tribes go to war for the ear of swing voters. People get increasingly angry and eventually violent language is used by people to describe their political enemies. The words "treason" and "traitor" get used. This is what happened to President Obama, and it is increasingly happening all over the world in liberal democracies.

Adversarial politics has always existed in liberal democracies. The problem is that it has gotten worse.

Corruption is also a problem. And I'm not talking here about underhanded, secret deals. I'm talking about legal influence peddling. Big business has bought politicians by providing them with the financial means of running a successful campaign in exchange for policies that benefit them, whether it be adjusting business laws, or cutting taxes or giving preferential treatment to specific businesses when issuing government contracts.

So the modern politician - the person who ends up in parliament or congress - is nowhere near an adequate representation for the people who voted for him/her. Ambition + money + propaganda is what happens. Yes, it has always happened to some extent, but our current situation is worse than ever.

What is needed are politicians:
    * who are not backed by rich benefactors
    * who are not engaged in defaming those who they disagree with
    * who prefer co-operation instead of competition
    * who see public service as service, rather than a chance at power
    * who are rewarded for being careful, judicious and objective
    * who depend upon experts for making hard decisions

Sortition is a way of doing this. By randomly selecting people to serve in parliament / congress, rather than electing someone from an established political party, modern democracies can be changed for the better.

Consider the following.

A randomly selected person serving in politics:
    * Is not backed by rich benefactors.
    * Has no incentive to defame those they disagree with.
    * Has not gotten to their position through competition.
    * Is given a chance for public service.
    * Realises that important decisions require care, and objective thinking.
    * Needs experts to communicate to them when it comes to making hard decisions.

Will this be a perfect system? Of course not. Once a person has won the electoral lottery, they may be approached and corrupted by big money and influence. This can't be helped but it can be mitigated, especially if there are laws set up that punish this crime and rules set up to prevent it. The question is not whether this system is perfect (it won't be), the question is whether this system will be better than what we currently have.

There are obviously a number of questions that arise out of this proposal. I've tried to anticipate them and I have written them here. Note that when I use the word "legislature", I am referring to the generic term that countries name as "parliament" or "congress" or "duma" or "diet" or whatever:

Q. Won't the legislature collapse if it was suddenly filled with randomly selected people with no experience now running things?
A. It probably will, which is why it needs to be introduced gradually into the already existing political system.

Q. How would this gradual change take place?
A. People would be selected randomly to serve in the legislature on a continual basis, rather than all at once. This would mean one or more people entering at least every month or so. As each new person enters, a person already in the legislatures would be forced to retire. A "transitional legislature" would exist in which the last of the elected politicians are gradually replaced by randomly chosen people. Once the last elected politician is retired, then the legislature would only ever consist of people selected randomly.

Q. How long would a person serve in the legislature? Are there term limits?
A. This is a good question and one in which there is no definite answer. There should be a term limit, but it should not be too short, and nor should it be too long. I would suggest somewhere between 6-12 years. When their term finishes, they are forced to retire and the only way they can get back into the legislature is if they are selected randomly again.

Q. What would happen if weird/crazy people end up being selected?
A. Well there are some weird/crazy people who are currently serving as politicians so the current situation is just as bad! One of the important things to understand about a legislature being populated by randomly selected people is that its size must be big enough to prevent the weird and crazy people from having too much power. This is about risk assessment, namely that the smaller the legislature, the less chance there is of a crazy person serving, but a greater chance of problems if by chance one does end up serving. In a larger legislature, there are greater chances of selecting a weird/crazy person, but a lower chance of that person causing problems. Thus a legislature needs to be of a certain size (probably more than 50 people) in order for this to work.

Q. Are there people who would be prevented from being selected?
A. There would be people who would naturally be prevented from serving. These would include children, people with debilitating brain diseases and conditions, people in jail, people living overseas, people who are too old, and people who aren't citizens (a person would have to prove citizenship before being part of the selection process anyway). Other criteria would include people who have violent criminal histories. There could also be added criteria, such as level of educational attainment and areas of study, but these can be determined later on.

Q. What if a person doesn't want to serve?
A. No one will be forced to serve. If a person refuses an offer to serve after being randomly selected, another person will be selected randomly instead and the original person would not be punished.

Q. What if a person is selected but doesn't really know how things run? What if they're called on to make a decision about something they don't know anything about?
A. This is a very important question. The fact is that many of our current politicians make decisions about things that they are not experts about. What are needed are experts who have the ability to explain important things to politicians, to help them make the decision. Governments all over the world have an entity called "the public service" in which a multitude of experts run various government departments. Once a person is selected and enters the legislature, they will be appointed a group of experts and advisors who can not only help them make make a careful policy decision, but who can also help them through the legislative process.

Q. Wouldn't having a randomly selected president be a bad idea?
A. Yes it would, but that's not what's being proposed here. This proposal is only aimed at the legislature, the group of people who create laws for their nation or state, not the executive. The role of the executive (ie president, prime minister) might be impacted by this proposal, but the person filling that position should not be selected randomly. This is because the role of an executive (who has control over the day to day functioning of the government) needs to be given to someone who is suited to that position.

(Edit 2018-09-12: If you don't understand the constitutional difference between the Legislature, the Executive and the Judiciary, you probably need to read this wikipedia article on the separation of powers. For Australian and British readers, you also need to understand The Westminster System.)

Q. Could random selection have any role in appointing people as president or prime minister?
A. The only way this could occur is to randomly select a person from a pool of qualified candidates. This would prevent unqualified people from serving in an executive role while still allowing some random selection. It would work better if the executive was a group of people, which is what happens in Switzerland. Another solution could be the executive being chosen by a group of people who were randomly selected for the express purpose of choosing the executive. Otherwise, the person could be selected via popular election (with all its problems). Note that the role of the executive is a separate issue to this whole proposal, which is aimed at the legislature (the law makers).

Q. Could random selection have any role in appointing people as judges in the judiciary?
A. Yes it could. As above with the executive, the best way would probably be to randomly select people from a pool of qualified candidates, with term limits applying.

(edit 2018-11-13)
Q. What process is used to randomly select someone?
 Random selection can be achieved in different ways. The most obvious modern day process would be people's names in a computer, and then software used to select someone randomly. This is a very simple process - it can even be done on modern-day spreadsheets.

Other ways would be a "lotto ball" system in which people's names are placed in plastic balls that are then drawn from a rotating cylinder - the same way lotto winners are determined, and, if you go back in history, how some people were chosen to serve in the Vietnam War.

No matter which method of random selection is used, the process itself needs to be fair and free from influence or corruption. The current process for modern elections - to preserve fairness, to count votes accurately and stop voter fraud - is a much more complex and corruptible system than random selection. The point here being that if we alread accept the risks associated with fraudulent elections by having a system that polices it, then there really is no problem in accepting the theoretical risk of someone fixing or manipulating a system of random selection.



2018-08-28

Which Celebrity has let me down the most? Bono.

(This is a copy and paste of what I wrote on Reddit:)
https://en.wikipedia.org/wiki/File:U2_21081983_01_800b.jpg

 I'm 49. I was a U2 fan before Joshua Tree came out. U2's music in the 80s was very different from the mainstream pop. They sang about important things. Bono used his position to fight issues like racism, hunger and war. Then, when the 90s came around, they effectively re-invented their musical style. They also poked fun at themselves.

 I have the U2 Time magazine in a box somewhere, along with newspaper articles and vinyl records and singles from the time. They were my favourite band.

But there were three things which made me lose my respect from Bono and U2. This has resulted in me no longer listening to them and feeling very let down.

The first was the Iraq War.

U2 wrote a song called "Bullet the Blue Sky" in which they criticised US involvement in Nicaragua in the 1980s. The song even got banned from US radio. But by the time of the Iraq War, U2 were reaping considerable rewards from being a mainstream musical act. They could have used their power and influence to oppose the war. It would have lost them fans and money, and the right wing media would have excoriated them, but it would have meant that they were consistent to their principles. But what happened? They remained silent. Bono even went to the White House and shook hands with Bush. Meanwhile The Dixie Chicks stood up for what was right and were the musicians that took the brunt of right wing anger. And then when it came out that the US was torturing people in CIA secret prisons and in Abu Ghraib prison, did U2, supporter of Amnesty International and writer of "Mothers of the Disappeared" do anything? No. For me, this was the loss of U2's moral authority.

The second thing that happened was the U2 Tower.

Yeah, you read that right. Property developers in Dublin approached U2 with plans to create the tallest skyscraper in Ireland. At the top of the building would be a "pod" where U2 would live and record music. Of course the building was never built - the global financial crisis killed off those plans. But U2 acquiesced to the original plans and provided some investment.

The third thing was their failure to pay taxes.

After the GFC Ireland was in a mess. Unemployment was massive and the government had a huge budget deficit. The government decided to cut spending on welfare and thus the people who were hit the worst were the poor. U2 had stashed a lot of its earnings in overseas accounts and investments - completely legal but it was the sort of thing that a lot of rich people had done. As a result, U2 were not paying in tax what they should've been paying. Of course Ireland's fiscal nightmare wouldn't have been solved by U2 paying more tax, but it was the ethics of the situation that mattered. By keeping their earnings away from the taxman, U2 were part of the problem rather than being a solution - and the people who suffered from this problem were the poor, the very people that U2 had spent its life publicly championing. U2's manager, Paul McGuinness, defended the actions of the band, first arguing that it was legal (which it was, but that's not the point), before then launching into what I call the "CEO defence", namely that he spent time saying how wonderful U2 was to the economy and how many people were employed by them over the years, which was just another way of saying "you peasants should be grateful".

Most Irish people think U2 are a bunch of prats. I agree. They, especially Bono, are hypocrites, the worst sort of self-serving celebrity social champions.

2018-08-01

Potential Genocide Perpetrator

As this week’s end, it seems likely that it is quite near time for killing those involved in the multiple and clearly delineated attempts to stage a coup d’├ętat against the legitimately elected Trump government and thereby kill our republic.

American patriots have so far, praise God, been remarkably disciplined in not responding to tyranny and violence with violence. For now they must remain so, armed but steady. But the time for such patience is fast slipping away; indeed, that patience is quickly becoming an obviously rank and self-destructive foolishness. If Trump does not act soon to erase the above noted tyranny and tyrants, the armed citizenry must step in and eliminate them.

It is, of course, far better if Trump does so, and I pray and believe he will. That said, the sheer, nay, utter joy and satisfaction to be derived from beholding great piles of dead U.S.-citizen tyrants is not one that will be missed if Trump does not soon do the necessary to save the republic.
Michael Scheuer, Trump supporter and potential genocide perpetrator.

2018-07-28

A solution to US policing

America has a very serious problem with bad policemen. It's unique in the Western world. Other industrialised countries just don't have the same problem. It's not as though other nations are perfect, and I say this because someone will think I'm saying that. And I'll also say that the US has some very good police officers and police departments.

It's the structure of US policing that is the problem. You have police and a court system at all three levels of government - federal, state and local/municipal. Not to mention university and college law enforcement (in Australia, where I come from, universities employ security guards, but policing in educational institutions are done by the state police as well)
Other countries just don't have this. In the UK, the police are overseen by a single government. They are broken up into various metro areas but the quality of training and standard of behaviour is uniform throughout the entire UK.

In Australia, the "police" are run by the state governments. We have a few federal agencies to help with law enforcement, but no local government has its own court system or police officers.

The US system is so bad because of local police. Because each local government area (cities, counties, parishes, towns etc) has its own police, it can only afford what it can get in local taxes. Thus poorer areas have low quality police and rich areas have high quality police... and the rich areas are corrupted by the local wealth.
The solution would be to have
  • A Federally funded and trained police force,
  • Controlled by the State governments,
  • Appointed to Local areas by their consent.
This would free up the costs of law enforcement for local and state governments by diverting it to the federal government. It would allow states a measure of control for states rights purposes, and allow local areas to decide whether or not they want their own cops or whether they want to save money by letting the states run their law enforcement.
Note: This is not a perfect solution. Corruption and incompetence will not be completely erased in this system. But it will definitely make things better. And I'm annoyed that I have to say this because idiots will come out and say "but there's police problems in UK and Australia too" and link to wikipedia articles.

(crossposted from Reddit)

2018-04-20

The U.K. Just Went 55 Hours Without Using Coal for the First Time in History

Bloomberg:

No coal was used for power generation by stations in the U.K. between 10:25 p.m. in London on Monday until 5:10 a.m. on Thursday, according to grid data compiled by Bloomberg. At the same time wind turbines produced more power.
This is good news. But it also shows further development is needed. The latitude of the UK places it in a zone which is rich in wind, but not necessarily rich in solar potential.

This means that decision makers should invest more resources into wind farms, especially those off shore.

It also means that energy storage facilities must be built. Wind (and solar) are intermittent power sources. The recent wind power increase is due to very high winds and will obviously decrease once the low pressure system disappears. Energy storage infrastructure is essential for the future development of renewable energy, as it can store the excess energy generated during peak production times (windy days for wind turbines, sunny days for solar) for later use during peak consumption times. This infrastructure can include the sort of lithium-ion battery storage that Tesla built in South Australia recently, and can be built on land already used by coal/gas power stations (since the distribution infrastructure is already in place).

2018-04-05

Evangelicals = ISIS?

This right-wing Christian movement is fundamentally anti-democratic. Their "prayer warriors" do not believe that secular laws apply to them, thus making it acceptable, if not honorable, to deceive non-believers in order to do God's work. Many evangelicals in the Christian nationalist or "dominionist" wing of the movement want the United States to be a theocracy. In some ways, this subset of the evangelical population resembles an American-style Taliban or ISIS, restrained (so far) only by the Constitution.

Have Christian nationalists staged a “soft coup,” with Trump as their figurehead?

Slightly overblown, but generally accurate in the view of this non-American evangelical.

2017-11-24

US per capita economic growth has collapsed since 2008

This is hardly a surprise for those who have been looking at economic stats. Above is a graph I have done based on a bunch of stats available from FRED. It is Gross Domestic Income, adjusted by the GDP Deflator, and divided by population to give a per capita result. Then the growth is averaged out over ten years. All figures are quarterly.

Gross Domestic Income
GDP Deflator
Population

So what you're seeing in the graph at each point (quarters of a particular year) is how the US economy has performed over the previous ten years. The downward lines reflect the various recessions that have hit since 1962 and broadly match the NBER definitions.

What is clear from the graph above is that the years following the 2008 recession have seen a deep and long-term collapse in per capita economic growth. Moreover, what has been experienced since 2008 has been unique in that the economic trough has not recovered - the only other comparable event was the 1973 oil crisis and its aftermath, which permanently capped per capita growth to around 30%, making the 1962-1973 period a "golden age" in comparison.

(Why GDI and not GDP? See Nalewaik)