There's something that's really worrying me about the current economic conditions: the combination of a drop in the CPI, the Fed funds rate going to zero, and the yield on Treasuries going to zero. Together, they indicate that deflation is not just a worry, it's already here. And once you enter a deflationary spiral, it becomes very difficult to get out of it.
Deflationary depressions don't just last a year or two and then you snap out of it, they can go on for 10 years or more. Actually, in most cases they seem permanent: they continue until there is a war that forces the country to spend or die. And during them, far more that the normal 80-90% of startups die. (In a presentation at UIUC, Max Levchin flippantly remarked "85% of startups die...that figure's been true throughout most of history, except during the Great Depression, when 100% of them died." That's not quite true - Disney and Hewlett-Packard were founded during the depression - but it's pretty close.)
1) Preventing a deflationary spiral is not that difficult. Declare a tax holiday, and print money to fund government operations. Keep printing until the price of land/gold/oil stabilize and start to rise again. At that point, ease off and return to a stable money supply.
2) Money supply deflation usually only lasts a couple years, four years at most. When the banks stop failing, the money supply stabilizes. Bernanke is hard wired to err on the side of inflation, rather than deflation. so I would not expect a deflation lasting more than a year or two.
3) Price deflation can last for decades, but that's a sign of a stable money supply and a rising productivity ( for instance from 1980 to 2000 in the computer industry). From 1870 to 1910 was a period of general deflation, but it was perhaps the best time for startups in the history of the world.
4) The New Deal had a devastating effect on small businesses. The NRA, AAA, OPA, and other alphabet agencies were insane. Here is a small taste: "The NRA was discovering it could not enforce its rules. Black markets grew up. Only the most violent police methods could procure enforcement. In Sidney's Hillman's garment industry the code authority employed enforcement police. They could enter a man's factory, send him out, line up his employees, subject to minute interrogation, take over his books on the instant. Night work was forbidden. Flying squadrons of these private coat-and-suit police went through district at night looking for men who were committing the crime of sewing together a pair of pants at night. But without those harsh methods many code authorities said there could be no compliance because the public was not back of it." source: http://mises.org/books/rooseveltmyth.pdf (p 45)
Want to try a startup with the code authorities breaking down your door to stop you from coding at night? And then there were was the crush of paper work, price regulations, etc: "A Michigan grocer who had run a successful business for 40 years testified that 'For the last six months I have been behind the counter ten hours a day, then up half the night filling out government forms. Sunday is needed for inventory reports, ration accounts or applications for coffee, sugar and canned goods. I couldn't keep up with it, so I closed my doors.' Small food distributors were going out of business by the tens of thousands a month." (p. 316)
Regarding 1), it's trickier than that. The issue is that there are two types of inflation, monetary inflation (more dollars floating around) and price inflation (stuff costs more money). Due to hoarding, these two types of inflation may not agree with each other. If you inflate the currency by printing money, much of it may be hoarded, since owning money is (in a deflationary environment) an investment. So the money supply increases, but consumer/investor prices continue to deflate!
Now you have large chunks of money hidden away, but prices continue to fall. This continues, until you reach a tipping point and price inflation returns. Then the hoarders spend their money, and prices inflate rapidly (price inflation catches up with money inflation). This isn't a good thing, since it's a major monetary shock, and can lead to a crisis of confidence, etc...
Probably the only way to avoid this scenario is to make sure that the newly printed money has a high velocity, but it isn't clear how to do this. The classical answer is to give it to lower income people, but it's not obvious that this will work in a deflationary environment (poor person buys TV, TV manufacturer hoards money).
You're absolutely right about the two types of inflation. In fact, any argument about inflation/deflation needs to start with throwing both words out the window. Let's use the world "dilution" to refer to increases in the money supply, and "contraction" to refer to decreases in the money supply. We'll use "price level increases" and "price level decreases" to refer to price changes, as reflected in the CPI.
If you inflate the currency by printing money, much of it may be hoarded, since owning money is (in a deflationary environment) an investment.
The government must print enough money to offset the money lost through the collapsing credit. Since the government is committing to keep the monetary base stable or increasing slightly, there is no incentive to hoard. The incentive to hoard only happens when their is a money supply contraction, not a price level fall.
A depression happens when there is a sharp contraction in the money supply due to a credit bubble collapse. Businesses that had grown addicted to credit, fail first. Consumers, expecting their debt to be diluted away, now have to save more and pay off their debts. Asset prices fall, causing a wealth effect drop in spending. The net is that fewer green bills are chasing the same number of goods, thus prices start dropping. Falling prices means less revenue for businesses. The hardest sectors hit are those that are in debt, as they borrowed assuming more money available to pay back the loans. But every sector is hit hard because of sticky wages, menu costs, etc. Businesses start freezing hiring and laying people off. The depression deepens.
The idea is that the Fed must inject money now to prevent the dominoes from falling. If the Fed prints enough money and gives it to consumers, consumers will be able to repair their balance sheets and quickly maintain or return to pre-crash spending levels. Thus all the contracts and wage agreements assuming the pre-crash money supply will be bearable, and no mass layoffs need happen. The crisis is averted
The big key is not too print too much money, or you risk triggering a hyperinflation. But by keeping a watchful eye on the hyperinflation canaries (gold, oil, and real estate) the Fed should be able to get it right.
(poor person buys TV, TV manufacturer hoards money).
That's fine - the point is to stop the TV manufacturer from having to make layoffs do to falling demand. In the real world, the manufacturer probably wouldn't hoard. Depending on their situation they might pay down debt, invest in expansion, or return dividends to shareholder.
2) In the presence of an activist monetary authority, like the Fed. As you mention in point 3, it can last for decades if nobody prints money. Even then, it's not certain: Japan's deflation has lasted for 18 years and counting.
3) The period from 1873-1896 was a terrible time for startups. It was a war of attrition, where the firms with the most capital just sat and waited for undercapitalized firms to fail. This is how Rockefeller and Carnegie made their fortunes: they bought out all their competitors cheaply when falling prices meant that their competitors could no longer compete.
3) What are your sources? I had read some stats about small businesses in that era, and it was stunning how much more vibrant it was than today. I wish I could dig them up, but I can't find them at the moment. But if you just wander around an old industrial city, you can see the buildings of the tens of thousands of small manufacturing start ups that dominated the era. The Carnegie example doesn't really help you - it was a startup founded in the 1870's. It was the Google of its era. Does Google's search monopoly prove that 1998-2008 was a terrible time for web startups?
But isn't an inflationary spiral / hyperinflation just as bad as a deflationary spiral? That wipes out a country's currency as well as the life savings of people.
In general, gross imbalances in the money supply vs. goods is not a desirable condition.
It's easier to control. The cure for hyperinflation is to just keep raising interest rates and buying currency in open-market operations. Eventually, the supply of currency becomes tight enough that firms can no longer raise their prices without losing customers. It's painful (just look at 1980-1982), but it works.
When countries don't do this, it's usually not because they can't or because they don't know how, but because there's another factor that will threaten the stability of the state if they do. For example, if a country has a large standing army, they can't shut off payments to those soldiers without risking revolution. Or if there's a large foreign-denominated debt, they can't stop payments without inviting war with the creditor nation.
In past comment threads, I've gone looking for various instances of hyperinflation, and they all fell into one of the two categories above. I'd welcome counterexamples, but nobody's provided any yet.
[Edit: Interesting, I found one. Bolivia in 1985. Like every other instance of hyperinflation I know of, it was caused by the government printing money to support expenditures that its tax base could not justify. However, in this case the expenditures were non-essential (things like subsidized gasoline), and so when the government eliminated the pork and committed itself to a balanced budget, the hyperinflation ended.]
The problem with deflation is that interest rates can't fall below 0%, and there's no guarantee that infusions of currency will reach the people that will actually spend them. Once rates are at 0%, banks and savers will just hoard currency instead of circulating it, so even if there're lots of unemployed people that desperately need the money, it won't get to them. "Pushing on a string."
I think that Bernanke's "helicopter" comment may have proved prophetic, and it may shortly become his best option.
Historically however, it seems that hyperinflation has been the more prevalent cause of economic woes, e.g. currency collapse, loss of people's life savings, etc. than hyperdeflation.
There doesn't appear to be an upper limit to hyperinflation (e.g. trillions of Zimbabwe dollars for a loaf of bread) whereas oil even in a deflationary environment shouldn't go below 0.
Personally, I am worried about the hyperinflation scenario and hope the government doesn't cause an overcorrection by its recent measures.
Hyperinflation causes more visible woes, because it's pretty obviously painful when your cash is worth 1/10th what it was yesterday. Deflation only causes that sort of visible pain when there're mass layoffs and bank failures, like in the Great Depression.
However, in terms of lost opportunity cost, deflation is far worse. It encourages people not to invest, because they earn better returns on their cash than on productive projects. So the society stagnates.
Think of the Dark Ages - a whole millenia lost, where nothing much happened. From everything I've read, it wasn't really a time of misery - people had enough to eat, they survived. But nothing much happened. If the Roman Empire had continued, even with its inflationary tendencies, we might've seen the industrial revolution a millenia earlier, and be writing these comments in 1008 instead of 2008.
Yummyfajitas has a good answer to this elsewhere on-thread. Basically, money collects at whichever firm needs it least. If you give it to low-income people and they spend it, then it's now in the hands of someone who probably just received a lot of money. If they don't need to spend it and expect its value to go up, they'll hoard it.
You need consistent inflationary expectations across the economy to keep money flowing. Keynes actually suggested that this is why it took a couple thousand years after the invention of money for a modern industrial society to take hold. A manor lord has every incentive to hoard the rents his peasants give him, so instead of pumping them back into the economy by funding the promising waterwheel a bright upstart proposed, he sits on his treasure hoard.
Hyperinflation was a recurring problem in China from 1100 to 1450 when the government used paper currency - http://goldnews.bullionvault.com/china_paper_money_chinese_p... I don't think that the Chinese case fits either of your two categories. The Chinese situation is actually quite analogous to the U.S. The government began to rely on fiat currency and inflation to fund its operations, and the temptation to keep printing money was too great.
Endowments have charters which limit the amount of their capital that can go to high risk investments - in other words venture fund and private equity firms. This means that they need to "rebalance" their portfolios and shift money that would normally have gone into venture funds back into public markets.
Am I understanding this correctly. They have a charter that determines the balance. This means that is a certain mandated class losses value, they have to top it up with funds from the other classes?
Deflationary depressions don't just last a year or two and then you snap out of it, they can go on for 10 years or more. Actually, in most cases they seem permanent: they continue until there is a war that forces the country to spend or die. And during them, far more that the normal 80-90% of startups die. (In a presentation at UIUC, Max Levchin flippantly remarked "85% of startups die...that figure's been true throughout most of history, except during the Great Depression, when 100% of them died." That's not quite true - Disney and Hewlett-Packard were founded during the depression - but it's pretty close.)
Some interesting reading: http://chronicle.com/temp/reprint.php?id=477k3d8mh2wmtpc4b6h...