Monday, October 27, 2008
Political Diary: Examining the Fundamentals
I and a number of my friends have been arguing about Obama and McCain's financial packages ... one of them even asked, "So, which of these guys is going to win the war on pandering?"
In sum, looking at the economy as a whole:
- Spending equals income. The nation's net income - the GDP - sets the limit on the nation's net spending. The GDP is made up out of what consumers buy, governments spend, investors invest, plus our net exports (that last bit is a current net negative). This has a few corolloraries:
- If government spending goes down, the GDP goes down, and hence, so does the aggregate income of the nation. Put another way, if you lay off a cop, he buys less donuts.
- If government taxes go down, the GDP goes up, and hence, so does the aggregate income of the nation. Put another way, if you cut the cop's taxes by 10%, he buys more donuts.
- We consume less than we earn. If wages double, spending doesn't double - it goes up somewhat less. This doesn't hold for transient events like stimulus checks, which people tend to "discount" as a one time event, but it does hold for "permanent" tax cuts, which people tend to factor in to their long-term plans. If you measure this factor over the long term, it's roughly sixty percent or so, depending on how you measure it. Take that as an approximation.
- Spending changes have a multiplier effect. If the government cuts spending or raises taxes, there's a disproportionate effect that ripples through the economy. When you cut the cop's pay by 10%, he buys 6% less donuts (and other stuff), which leads the donut shop cutting its employees' hours 6%, which leads to 3.6% less other expenditures ... in total, that 10% becomes roughly a 25% hit on the economy. A tax cut of the same magnitude has a positive effect, also about 2.5x its initial value. This is also an approximation.
- Lack of liquidity increases the multiplier. If your income goes up 10%, but all of your money is tied up in long-term CDs, you can't buy that new house. If mortgages are not available, you can't buy that new house. If you don't have a credit card, you have to wait until after Christmas to buy gifts, even if your Christmas bonus is a sure thing.
The actual economic picture is MUCH more complicated than this Keynesian ideal (for example, net exports have an opposite effect, and this does not factor in investor confidence, etc), but at the 50,000 foot view, it seems like a few corolloraries in this crisis might be:
- Government services should be maintained. If the government does a radical belt-tightening or even a spending freeze relative to its current growth, there will be money in the government coffers that does not go out to the economy - and all those defense contractors, cops, social workers, civil engineers and so on that are out of work are not going to have money to spend or save, having a further impact on the economy.
- Taxes should be cut. If the government raises taxes, even for the "good" reason of being fiscally responsibile, individuals will have less money to spend or save, having a further impact on the economy.
So, shockingly, the two candidate's plans to cut taxes are not as horrible as I thought they were. I'm not completely sure about the above logic, as the economy is very complicated, but while I'm speculating I'm going to go further on a limb and try a few more ideas:
- If (if!) taxes are to be raised, perhaps they should be raised on the high end of the consumption curve. I'm not certain, but I strongly suspect that the consumption curve above is nonlinear - that the higher end of the curve flattens out, as Bill Gates and Warren Buffet do not have a million homes each. Since most of the money is on the high income end of the curve anyway, this is where you will have to raise taxes. It will hurt - don't kid yourself that it won't hurt - but it will have less impact than if you tried to get the same money out of the bottom end of the curve. (For example, at the bottom end of the curve people will go into debt rather than starve).
- A tax on consumption would directly impact the consumption curve, and should be avoided. Taxing consumption is like putting a brake on the relationship between national income and national spending - it will decrease the marginal rate of consumption (the 0.6 above) and raise the Keynesian multiplier (the 2.5 above) putting the brakes on the economy. This seems to indicate that sales taxes are a bad idea, and perhaps is bad news for the FairTax idea - the FairTax may be designed to make the system more fair, but it applies taxation to a place in the economy that will interfere with the growth of GDP. I haven't dug into the FairTax in detail, so they may have an answer which will ease my fears on this issue.
I'm not arguing here based on whether I think government should be big or small, or whether I want higher or lower taxes. I want lower taxes so I can buy more books, pay off my house more quickly, or get a new hybrid electric car; but more importantly, I want lower taxes because I think it will get the economy moving, as I've stated above. Speaking frankly, I have no desire to soak the rich: the rich pay most of our taxes (no, really) and do most of our investment upon which many of us rely for their jobs. So I don't want to make their lives harder; however, the rich have a greater ability to pay and still be able to invest and consume so if (if!) we were to raise taxes they should go up on the higher end. So based on these ideas ...
- Obama's tax cut idea is not horrible. It raises taxes on wealthy people, but overall it is a tax cut, and should stimulate the economy.
- McCain's tax cut idea is better. It lowers taxes on more people, for a larger overall total, and should stimulate the economy more.
- McCain's spending freeze is questionable. It's not a terrible idea, but it does suggest that we should focus on trying to grow the economy to the point it can pay for our services, rather than cut our services and thus stop pumping that money back into the economy.
- Obama's idea to back up states the way we are banks is not a horrible one. The biggest problem is the moral hazard it creates ... obviously all 50 states cannot do this forever, or basically we're draining our income taxes into our state budges. But if we allow governments to stop providing services, it will hurt the economy.
A few notes on things not completely covered by this model:
- McCain's mortgage plan is ... WTF? Seriously, as a friend put it recently, "if we're going to spend 700B we should think out of the box". The answer may or may not be McCain's plan, which is more complicated than my simple little model and my dime-store understanding of economics. It would save homeowners and reduce risk to lenders, which should help things ... but the moral hazard, and investment risk, worry me.
- Paulson's plan to buy into banks is ... not so bad? The capital injection into the banks, because it comes with a 5% dividend, sounds like a good idea. If the government is buying things at a steal, this could help the taxpayers.
Unfortunately, I think we are in "uncharted waters", as both the Google CEO Eric Schmidt and Google gadlfy Mark Cuban. We have a crisis of investor confidence, which has led to a severe drop in the amount of investment recently. Some bold people with liquid cash, like Warren Buffet, may make money ... unless we are on the high side of a slide down to the market losing 90% of its value, as another friend pointed out happened in the Great Depression.
In this case, larger issues that the simple Keynesian model come into play:
- Moral hazard. What kinds of behaviors do our policies enable? As one friend. has repeatedly reminded me, "if this goes on" we would keep bailing people out forever. The solution to that is to increase regulation on the bailed-out industries ... which will slow them down. So do you not bail them out ... in which case the lack of a bailout will cause a certain amount of pain. How much pain is too much pain? If we avoid bailouts and regulation, is that saying that we as a nation have to depend entirely on the market and sort of hope that maybe it will do the things we need to supply our nation with the things it needs to defend itself and maintain its global position (which in turn enables the functioning of our market)?
- Investor confidence. As investor confidence drops, the amount of money invested in the economy drops ... causing multiplicative effects as we have seen above. In recent times these "factors" have become highly nonlinear and disruptive ... so we can't predict how much investment will happen from overall effects on the GDP. We have three big variables we as a nation can use to affect the economy: how much we tax, how much we spend, and how easy it is to lend money. The third factor has gone off the rails because of investor confidence ... it is looking like we have little control of this right now, though the TED spread is getting better. The tax and spend equation is also off the rails because of our huge debt, which means that some huge percent of our government income is not directly getting fed back into the economy in terms of goods and services, but instead is being paid off as debt, sometimes to people outside this country.
If we could wave a magic wand and fix things, we'd see a country with:
- No debt, and thus not driving with the parking brake on.
- If possible, less taxation overall, which would make it easier for people to spend/invest what they have.
- If possible, less government spending in relation to GDP, which will increase the rate of growth of the economy and over time give us the revenues we need to do a wide variety of useful government programs.
- Government spending close to government income, aaaaand...
- A large cash reserve. This would be used to smooth over the rough times when we need to drop taxes or expend services to jumpstart the economy. It could also be used for bailouts or to smooth over state budgets ... money that was loaned out at strict rules and/or with high interest. Which makes you wonder ...
- Is the idea of buying in to the banks a better one than we ever realized? Should the U.S. Government become a (safe, risk-averse) investor?
Right now I'm still researching, so these are more discussion points for ideas, not proposals at this point. But my big point here is I'm not talking ideology here. I'm not saying that big government is good or bad or saying that it would be wrong to redistribute wealth or wrong to encourage dependence on government. I'm saying, economically, some things will work and some things won't. Taxing the hell out of ourselves is going to slow the economy. Cutting government spending is going to slow the economy. Running with low taxes and high spending too long is going to raise debt and slow the overall growth rate of the economy, so when we get the economy running again we need to trim spending further to raise the investment rate - even if you want to ultimately provide more government services, you have to grow the economy first or you're trying to save the world with your hands tied.
If we want to have the resources available to us to do what we want - whether we want to buy guns or butter - we need to clean out the tank, release the parking brake, and drive the country's economy in the right direction.
Sunday, October 26, 2008
Emotional Memory and Adaptive Personalities
So a couple days ago I sent off the final draft of a paper on Emotional Memory and Adaptive Personalities to the Handbook of Synthetic Emotions and Sociable Robotics. Tonight I heard back from the publisher: they have everything that they need, and our paper should be published as part of the Handbook ... in early 2010.
"Emotional Memory and Adaptive Personalities" reports work on emotional agents supervised by my old professor Ashwin Ram at the Cognitive Computing Lab. He's been working on emotional robotics for over a decade, and it was in his lab that I developed my conviction that emotions serve a functional role in agents, and that to develop an emotional agent you should not start with trying to fake the desired behavior, but instead by analyzing psychological models of emotion and then using those findings to design models for agent control that will produce that behavior "naturally". This paper explains that approach and provides two examples of it in practice: the first was work done by myself on agents that learn from emotional events, and the second was work by Manish Mehta on making the personalities of more agents stay stable even after learning.
The takehome, however, is that this is the first scientific paper that I've gotten published in the past seven years, and one of the few where I was actually the lead author (though certainly the paper wouldn't have happened without the hard work and guidance of my coauthors). Here's hoping this is the first of many more!
Saturday, October 25, 2008
Political Followup: The Bailout
A few followup thoughts on the bailout, after discussion with friends:
- Why we're in this mess:
One of my friends pointed out that the real problem was banks buying toxic debt and not being willing to sell it for pennies on the dollar; because they sold us a story that they're "too important to fail" we now have a "no banker left behind" bill. He suggested that the money should go instead to augment the short-term commercial paper market. Well, this is now happening. Maybe this guy should be my financial advisor. Another friend had a similar idea and said it reminded him of Hamilton's Bank of the United States, a 20-year experiment that was used to put the new country's fiscal system on the right track by reducing speculation and increasing private commercial investment. So there is precedent for these experiments.
- The Fate of Libertarianism:
The same friend was discussing this situation and said (in his perspective) that it was the death knell of the Libertarian party. His friend said "try living without the FDA" and pointed out the current food crisis in China as an example; he then suggested that the same thing could be said of the financial markets - perhaps the argument might be "try saving without the FDIC", or "try investing without the Fed". Maybe that could work, but you'll find plenty of libertarians who'll tell you we weren't doing pure libertarianism before (see the Community Reinvestment Act) and so it's pointless to blame libertarianism for this problem. Nevertheless, the blame game is going around, and it's been difficult for me to get to the facts because so many people examining this situation are hopelessly partisan. More in a bit.
- The Fate of Capitalism:
Another group of friends were wondering about the fate of unrestrained capitalism in the face of this situation. A fair number of Republicans agreed with them over the past few years, calling for increased regulation which didn't happen while Democrats were arguing for less regulation - an odd flip of their normal positions. George Bush is being accused of being a "big government socialist" for the bailout by conservatives and liberals are calling for more regulation. Another friend was wondering if this would lead to fundamental changes in the capitalist system, and, shortly thereafter, the U.S. government started buying shares in banks. If successful, these shares will jump-start the flow of credit and, if the bailout is successful, will win for the taxpayers. If unsuccessful, we will be in the Second Great Depression, and the loss of that $250B will be the least of our worries.
So where do I stand? Right now everyone's too partisan to think clearly: the Republicans are blaming the Democrats, the Democrats are blaming the capitalists, the libertarians are blaming the government, the bankers are looking at their shoes, and no-one's trying to look at the many Americans we were encouraging to buy homes and who are now defaulting. I don't know if everyone's so scared they're grabbing for their favorite bugaboos or whether, more cynically, they're trying to use the crisis to get their favorite policy change implemented - or perhaps both.
But what I do know is that the amount of the bailout - seven hundred billion dollars - is only around 5 percent of our current GDP. At the "worst" point of the Great Depression, 1933, the GDP fell to about 50% of its pre-Depression levels. So at worst, the bet Bernanke is making is 10% of the GDP we should expect to have if things go really south. So ask yourself: is it worth it to take 5% of your income for a year --- roughly $2300 from each American --- to invest in a business upon which you depend for 50% of your income? Of course, the question isn't that simple: what help is the $2300? what's the chance that the business will fail anyway? what's the chance you will get it back? and, really, what's the likelihood that the $2300 is going to cause your crazy uncle to do something stupid again?
So I think it is worth it as long as we do the following things with the bailout money:
- Make it accountable: We need to know where the money is going and why.
- Don't throw good money after bad: If a business is going to fail anyway, don't prop it up.
- Put the money where it is needed: Whether this is buying short-term commercial paper, propping up mortgages, buying in to central banks to give them liquidity, it should be something that really helps in a material way with our real problems.
- Don't give the money away: The stakes we are buying in banks pay a hefty dividend. It goes up if the banks don't pay us back. All our other investments should have similar incentives for the taxpayer to be paid back.
- Don't get the government in the business of business: The government encouraging banks to do things they ought not to was a key part of this mess; we don't want the government meddling in the running of these businesses because what government officials think that businesses "ought" to do often has little to do with reality of the market.
- Don't let business get away from its fiduciary duty: Business leaders and individuals taking on more risk than they should was another part of this mess. Part of the game of people running businesses as businesses without being exposed personally to the businesses losses is the fiduciary duty of the people running the business: they are responsible for being responsible with the money that they have been given. Smart regulations should be put into place to discourage financial institutions from taking on the huge risks that they did in this crisis, without taking away their flexibility to do what they need to do to keep the market moving.
In short, if we carefully use this money responsibly, we may be able to blunt the impact of this downturn; if we're even more careful in how we invest it, as we appear to have done by investing in the major banks, the taxpayers may even come out ahead. We're not out of the woods yet, but there is reason to be hopeful that the bailout can help if we hang on to our principles.
One more followup on the Time Capsule
So back in June I purchased a Time Capsule from Apple, a one-terabyte wireless hard drive. It's supposed to make backups painless, happening in the background over the network. Initially I had a lot of trouble getting it to work with my wireless network, having to hard-reset it so many times I eventually recommended "don't buy it". Sometime in September Apple updated the firmware, and I recommended that it was worth trying out.
I didn't follow up with any further recommendations because for a while, it just stopped working again. Well, this time it was not the Time Capsule's fault: the wireless network it was connected to was having difficulties. These were orthogonal to the earlier problems, in which the Time Capsule couldn't even see the network no matter what I did; this time, after restarting the wireless router, the Time Capsule has been working swimmingly.
So, really, now you can get yourself a Time Capsule. Seriously. It's already saved my butt more than once, having files backed up every hour that my laptop is on my home network; I've already used the Time Capsule to recover earlier versions of a variety of files, rather than recreate them. (I am obsessive-compulsive about saving many versions of a file, but having it happen automatically is even better than me manually having to remember to save a file with a new version number hanging off the back end - not that I've stopped doing that.)
Saturday, October 04, 2008
Political Diary: The Bailout
So the bailout of the financial system passed this week. And I breathed a sigh of relief, sort of, and wondered what I would have done if that was my decision to make. I'm torn by two conflicting feelings: first, that we cannot afford to let the financial system collapse, and second, that we cannot afford to throw good money after bad.
Sometimes, of course, we have to let things die. But the overall financial system is not one of them. Financial instability is a vicious cycle: once the system becomes unstable, investors become afraid to invest money, making it hard for people and businesses to get money, causing more failures and instability. Put another way, Main Street depends on the jobs provided by businesses that depend on Wall Street's ability to lend money.
As I understand it, currently many businesses large and small depend on short-term loans for everything from operating capital to pay for stock in their warehouses to emergency funds to pay for the roof busted in last month's storm. They get this money from banks, who in turn get that money from you and me; they pay these loans back with interest, which the banks shave off as their cut before paying you and me our interest.
But banks right now are afraid to lend money, because so many banks, funds and businesses have recently collapsed. On paper, the banks have enough assets to loan out, just the same way as the average American with a decent-priced home has enough assets to buy a Lamborghini. However, most Americans can't buy a Lamborghini at the drop of a hat, even if they wanted to, because their wealth is tied up in longterm IRAs or CDs or in stocks which have dropped or homes with heavy mortgages. SO on paper you have the money to spend on a midlife crisismobile, but in practice you don't.
The technical term for this is illiquidity, which is just a fancy way of saying that have wealth on paper, but can't turn it into money without incurring terrible losses. A lot of banks are in precisely this situation: they want to loan money to a potential new homeowner or to a businessman trying to repair his roof, but their money is all tied up in investments they can't move without losing big --- in this case, mortgage-backed securities.
These "mortgage backed securities" arose because over the past several years housing prices were going up and up, and the government encouraged banks to make loans to people who traditionally were at higher risk than normal. These so-called subprime mortgages were packaged up into "investment vehicles" and "sold off" --- essentially meaning the banks that made the loans sold the rights to the interest that the mortgages would make to other people. That was a good deal while the economy was great because people were paying their mortgages or trading up to better homes. Then the economy started to stall and the housing bubble burst ... and suddenly people aren't making those payments.
This led to a "liquidity crisis" which is just another big fancy word for "runs on the banks". Perfectly good banks with great track records and billions of dollars on their books that hit a few rocky patches suddenly saw their money dry up in a number of weeks, leaving them with large short-term debts which in theory they could easily pay off ... except they couldn't move their mortage-backed securities. One bank failed after the other and in the end banks got afraid to loan money to anyone.
So the point of the bailout: those mortgage backed securities are not actually worthless. If the economy gets back on its feet (which, eventually, it almost certainly will) and home prices start to rise again (which, eventually, they almost certainly will), some huge percentage of those securities will pay off. It's just a long waiting game, a game which banks can't play because they're constantly lending and spending, but which the government can play, because of its deeper pocketbooks and constant source of real income from taxpayers. So the theory behind the bailout is, the government will buy those illiquid securities, allowing Wall Street to start lending money again, so business owners can stay in business, and Main Street can keep its jobs.
Which goes to my second concern, throwing good money after bad. Our civilization has experienced Dark Ages. Our country has experienced a Great Depression. What if we go through ten or twenty years of economic depression, and most of those mortgage backed securities are essentially worthless? That's over two thousand dollars out of the pocket of every taxpayer, another $700B on top of our $10T debt. Or, worse, what if the problems in our economic system are more systemic, and other banks are about to fail for other reasons? We might need that $700B for something else.
So, as I understand it, the bailout is needed. But, as I understand it, the bailout is risky. So what would I do if I was in charge? Well, here are a few rough ideas:
- First, go before the American people, and explain in clear terms what we are doing. Main Street's jobs depend on Wall Street's ability to lend money to businesses, but that's crippled right now because Wall Street tied up a lot of money in mortgages Main Street is having trouble repaying in this weak economy after the housing bubble burst. If the government takes over those loans, banks can start lending money again to keep business doors open and paychecks flowing, ultimately paying off those loans so that the government (and the taxpayer) is paid off for its investment.
- Second, demand oversight and accountability. This is seven hundred billion dollars we're talking about here. That's a year and a half's worth of defense spending, or almost enough to pay for the fireworks at the opening ceremonies of the Beijing Olympics. The original version of this plan gave massive powers to Treasury Secretary Paulson and made his actions immune to court challenge. That's a non-starter --- we understand the need for swift action, but it should be open, transparent, and if necessary subject to legal challenges.
- Third, demand foresight and forebearance. A necessary step in this process is to not just review the needs of the banks with mortgage backed securities, but to look at the health of the entire financial industry and to make sure there are no other trees that are rotted to the core and about to fall down. If other big problems are discovered, Secretary Paulson should have the authority to delay or defer spending that money and should go back to Congress if need be to ensure he has the right to apply this money where needed.
- Fourth, think outside the box. Could this money be applied in other ways? Could we take $70B of the $700B and use it to help homeowners who face foreclosures? I understand that homeowners taking on responsibilities they weren't prepared to handle helped cause this, but the banks are also to blame for lending the money. If we make it possible for the homeowners to pay off the banks ... how does that hurt the banks?
Some people have decried the culture of greed and called to make sure that none of this money goes into CEO's parachutes. I empathize with that but ultimately don't understand it. The problem here wasn't greed --- it was risk. We want people to be "greedy" in the sense we want them to take actions that benefit themselves --- to make a profit, in short. But in business you have what's called a "fiduciary duty" to make sound decisions for your shareholders. Some of the people who are involved in this really fucked up --- be they homeowners who got too big for their britches or CEOs who chomped one too many expensive cigars. But many other people played by the rules and then got caught by extraordinary circumstances.
Like a response to a hurricane, we need to come in and help the survivors without recriminations --- and then make sure that when it is time to rebuild we don't encourage people to put themselves at risk.