The GI Bill and Reconversion.

Unemployment Before and After the War.

Figure 1

Much of the boom feeling of the war came from the fact that unemployment was as low as the unemployment rate had ever been (or ever would be).  Obviously much of the cause of this was because over the course of the war some 16 million men (and women) would be inducted into the army and large numbers of civilian we hired by the government in support roles.  However, many women, older men and younger men entered the workforce which offset to an extent some the decline in "typical" workers.  Figure 1 shows the change in labor force participation as a percent of the total population.

Figure 2

There was some concern that after the war these returning soldiers would come home to a retrenchment of the depression and swell the ranks of the post war unemployed.  The call for a Serviceman's Readjustment act of 1944 (The GI Bill)was in large part put together in an attempt to soften the return of soldiers both for the sake of the economy and for the soldiers themselves.

There are three main aspects of the GI Bill. First, unemployment benefits of $20 for the first year after a solider was decommissioned.  Second, there were provisions in the GI Bill that allowed returning soldiers to buy homes with no down payment, this will be dealt with in more detail in a later lecture.  Third, and most relevant here, was the education benefit/stipend that paid up to $500 in tuition costs and after an amendment to the law in 1945 also paid a living allowance of $65 dollars a month for unmarried veterans and $90 a month with an additional allowance for dependents.  They were raised once more in 1948 to $75 for a single veteran and $105 for married veterans with children.  For comparison, a factory worker could make about $200 a month in 1947.

The red line in figures 1 and 2 offer a fairly crude idea of the impact of these education benefits on the post war employment situation.  They suggest that the educational benefits reduced the labor force/population ratio by up to around 2.5% and reduced the unemployment rate by upwards of 4%.  While these numbers should be taken with a cup of salt they do offer some evidence about the impact of educational benefits, it helped soften the impact of the returning soldiers on the job market.  As well, from a Keynesian perspective they remained consumers and were the conduit for the government to funnel money into education institutions of all types.  The magnitude of the effect is obviously not clear, especially since the unemployment benefits of the GI Bill probably artificially increased unemployment (though there is still the Keynesian aspect of the unemployment payments).

GI Bill education outcomes for African-Americans and whites.


Turner and Bound offer a pretty good picture of the role that WWII and the GI Bill played in the educational attainment for both blacks and whites.  The picture they paint is pretty straight forward and perhaps not unexpected.  It is important to approach the regression results cautiously because the confidence intervals around the estimates are fairly high though the results are still worth thinking about.  At any rate, the summary tables offer us some insights.

TB Table 2

First, lets look at Table 2 from Turner and Bound.  These numbers are taken from a different survey than the regression results.  These numbers come from (as noted) the 1979 Survey of Veterans.  About 1.9 million african-americans served in the military during the war out of a total of 16.3 million total.  That works out to about 11% of the military.  Given that, african-americans are slightly underrepresented in this survey, making up less than 7%).

It should be right away obvious that african-americans are less educated going into the war.  As well, a slightly higher share of african-americans used GI benefits but they have far far lower college graduate rates.  In fact, 5 out of the 165 african-american men in the survey received bachelors degrees.  Years completed are also clearly lower.

TB Table 3

Table 3 summarizes data from the 1970 census, the pool from which Turner and Bound get their sample. Here the disparity is also clear.  More whites than blacks were high school graduates and ultimately far more whites, both veterans and nonveterans went to college.  However, white veterans were about twice as likely to to go to college than white nonveterans and black veterans were three times as likely to go to college if they had served in the military.  That number demands a little nuance, though.  First, while 6% is three times as large as 2% it remains dismal even compared to college graduate rates of nonveteran whites.  Secondly, and this goes for the difference between both white and black veteran and nonveterans is that many of those called up for the draft were declared unfit for service (IV-F) becuase they failed basic literacy tests.  African-Americans were rejected more often than whites on these grounds, as Turner and Bound point out in 1944 a third of african-americans were rejected becuase of basic literacy issues and less than ten percent of white draftees were.  So, one would expect those serving in the military, both black and white to be more educated on average than their general postulation and you would expect black soldiers to be even more educated on average, table 3 seems to support this.

TB Table 4 

Finally, table 4 from Turner and Bound divides up the african-american sample by region.  I should note here that the south is defined to include seven southern states with the most stringent segregation laws (the uses a broader definition of south in the appendix).  here the difference between veterans and nonveterans within each region  do not seem that different.  However, across the board education attainment is lower in the south than the north.

The regression results.

TB Table 5

Table 5 offers the simple regression results of Turner and Bound.  The regressions suggest two things.  1.  Serving in WWII and having access to GI Benefits increased the probability that a veteran would graduate college, it also meant he would stay in college longer than otherwise.  2.  This is true of all groups except black men born in the south.  It is important to bear in mind that these are mean effects.  That means that the increase in the probability of 3.5 percentage points is significantly "water down" by the fact that only around 2 million soldiers (or 1/8th of all WWII vets) went to college.  A little playing with the weights suggests that in fact the probability of finishing college if you start was increased by around 30%.  There are of course a number of things to unpack from that but it does suggest that the stipend had a pretty significant effect.  That should be evidence from the Atlshculer and Blumin reading which details a lot of the supply side responses to in the influx of veterans.

The effect is even more pronounced in the case of african-americans, atleast in the north.  They show a higher mean effect, while at the same time a smaller share of african-american veterans attended college after the war.  Again, though, that needs to be unpacked because higher education did not respond as flexibly for the benefit of african-americans as they did for white veterans and so far more african-americans were denied even entry into higher education for lack of supply.  This brings me to the big takeaway from these readings.  Segregation is essentially a constraint on supply..  We will see this phenomenon in spades when we talk about housing.  In terms of housing, "ghettos" were not exactly natural phenomenon, they were basically created as clearly delineated spaces where african-americans were tolerated.  African-americans were then crammed into these spaces with the perverse result that housing in run down ghettos (because of a lack of access to credit to make improvements) was more expensive than in comparable (though nicer) white areas of cities.      

They effect on college educational outcomes for blacks is similar.  The pool of colleges available to african-americans was very small and not very well funded and so there was an inelastic supply response (though enrollments at black colleges more than doubled with assistance from the federal government) to what could have been a very significant increase in demand.   

Non-collegiate education under the GI Bill.

As Turner and Bound point out, college education is only a small part of the GI Bill education story.  While around 2 million verterans went to college using the GI BIll about 5.6 enrolled in noncollege education programs.  Altchuler and Blumin list the other programs: "3.5 million [attended] public and for-profit schhol; 1.4 million in nonagricultural on-the-job training; and 700,000 in on-the-farm training--5.6 million veterans in all..." (page 151).  This spawned a boom in subcollege schooling, much of it fraudulent or of little clear value.  However, while there were plenty of scams such as the "National Chicken Sexing School" the GI Bill also gave birth to what would become the Cullinary Institute of America.  There was also perhaps a misallocation of training for the jobs that were avalible and important to the post war economy.  Altchuler and Blumin point out that there were 30,000 students enrolled in tv and radio repair course in 1949.  However, this seeming over supply also seems to suggest that man verterans were able to get training for jobs that kept them out of factories and got them into skilled jobs.  All in all, though, the legacy of subcollege for-profit education after the war has not been given serious enough attention.  The tone of Altchuler and blumin, however, makes on think of the contemporary for-profit college industry and their relationship to veterans.

Breaking Windows.

The broken widow debate is raging once more through the blogosphere Dan Kuehen points us to Robert Fellner who offers in Dan Kuehen words "Higgs's critique of government spending". its reminded me of why the Higgs critique doesn't hold much water. Also, while I'm exercising some pet peeves about the Higgsian conception of GDP I would like to discuss some nuances of the Broken Window/Alien Invasion/WWII debate.

I've spent a lot of time with Higgs's thinking about GDP.  Much of his argument centers around an argument of authority invoking Simon Kuznets.  The problem is that Kuznets never really figured out how to deal with government in GDP accounting becuase it was developed before WWII and the rise of the "welfare state" proper.  In Kuznet's conception GDP was a kind of welfare accounting and that goods and services were categorized in terms whether they were welfare inducing now (C) or welfare inducing later (I).  However, some goods defied this simple categorization, in particular munitions   The issue was how do you characterize something like munitions?  If it is not welfare enhancing today or tomorrow, the argument runs, then they must be intermediate goods and thus not part of GDP*.  But of course you now run into the problem that you are arbitrarily assigning goods as intermediate and final goods based on whether you think they are welfare enhancing or not.  The solution to the problem is to adopt the approach the BEA eventually did, which is simply trace the production process of goods to their "end purchaser" and call those final goods.  However, this means that GDP is not a measure of welfare, it is an accounting identity.

Now, to the extent that one is tempted to use NIPA as a proxy for welfare the Higgsian view should be criticized because it ignores the flip side of national product: national income.  If the government pays someone to fix a broken window or dig a trench, that person's welfare has been increased regardless of whether he has produced anything strictly welfare enhancing for someone else (I'll come back to the net effect in a second).

The second prong in the Higgsian critique is that government purchases do not take place at market prices.

Here I'll let Robert Fellner explain it:

To see why, we first have to understand why spending at all can be considered a measure of wealth. Upon reflecting on this matter, we realize it has to do with prices (specifically market prices) and the information that they convey. Whereas one finds his lot improved by the purchase of 5 wheelbarrows at the market price of 10 dollars a wheelbarrow, we can conclude the individual, and thus the economy as a whole, is wealthier to the tune of the utility that 50 dollars in spending has granted him, specifically the subjective value of the additional 5 wheelbarrows. Moreover the spending of 50 dollars represents the creation of these 5 wheelbarrows and of course the additional utility they grant, hence why they were purchased. This is the key. Wealth is not measured simply by the dollar amount of spending. Wealth comes from the goods and services provided in exchange for money. This is why spending matters and is accurate as an indicator of wealth. If any part of this process is diluted, the quality of spending as an accurate indicator of wealth diminishes greatly. 

So why does consumer spending work well in this regard, where government spending fails? The answer lies in the prices. In a free market, all participants are subject to the profit and loss test. Namely, if one consistently spends more than he earns, he eventually becomes bankrupt and removed from the market altogether. In order to prevent this "death by free-market" one must learn to generate a profit; which is done by allocating resources efficiently. As all market participants engage with one another in this task, prices emerge for all the various goods and services within the economy that reflect their valuation to the economy as a whole (the price of course being derived both from the subjective valuation of the good as measured against the scarcity of the good, put more simply: supply versus demand).

This ignores three basic things.  First, it ignores the work of Alfred Chandler (And J.K. Galbraith, btw) who has in great detail explained how large firms in certain industries vertically and horizontally integrate specifically  to avoid having their production process (and prices) dictated by "the market" but rather prices are determined by negotiations between divisions of the firm.  In similar fashion, the government does not simply dictate prices but rather negotiates them with firms (when they have some degree of monopsony power at least).  Is negotiation really a less efficient means of price discovery than anonymous markets?  Are anonymous markets really the dominant free market structure anyway?  I'm less inclined to assume no to both of the questions than an Austrian is.  Anyway, I'm not denying gold plating but I am suggesting that the price the government pays is not a priori less arbitrary than the price early adopters pay for an Iphone (to illustrate this from a different angle).  


That may have been a clumsy segue but there is a more fundamental problem with the idea that "market prices" capture welfare better than government prices.  Even assuming all C and I are allocated by perfectly competitive markets competitive prices only capture the exact utility produced by the good for the last (marginal) buyer.  Everyone else in the market is accruing either a consumer surplus or a producer surplus.  To capture that involves a whole series of arbitrary assumptions that are impossible to make and best avoided taking NIPA for what it is.


Again, though, if we look at income it may be easier to make welfare statements.  If the government buys a loaf of bread from the baker it seems easier to make welfare statements (though I still advise against it even if I am as prone to do it as everyone else).  The baker's welfare is unambiguously higher if the government (or anyone) pays $100 for his loaf of bread instead of $1.  Unlike the troubles you run into with trying to make wealth/utility statements about who purchases what for who and why we can make a relatively uncontroversial statement that more income is better.


Okay, so now on to net effects.  There are definitely certain conditions under which the broken window falacy is wrong. The increase in GDP from fixing a broken window from one of the many curiosities of NIPA accounting that, again, underscore why you shouldn't act like GDP literally means total wealth, utility or welfare.  If I grow my own wheat and bake my own bread it doesn't count as GDP but if I buy it from the baker it does despite the fact that the same "utility" or wealth is created when I make my own bread (ignoring issues of scale efficencies***).  Likewise, fixing a broken window shows up as GDP even though I as an individual would be better off if the market transaction never happened and I spend my broken window money on components of GDP I  actually want to spend them on.


However, if the government borrows money to fix my window there are conditions (that aren't that ridiculous to meet and where the WWII example is instructive) where there can a net increase in GDP as income as welfare.  The relevant variable here is the Debt to GDP ratio (not the absolute level of debt, so take your Ricardain equivalence somewhere else) which allows the broken widow theory to work because there is no full redistribution through taxes.  The government can borrow today, give me the money to fix my window and then sit on the debt for as long as it needs to and let the burden of that debt shrink (under certain growth/interest rate conditions) away as nominal GDP growth increases, paying only the cost to service the debt as it shrinks.  This spending, by the way, also increases nominal GDP grow at the point in time that I pay for my new window assuming no crowding out it will directly off set itself and there will be no increase in Debt/GDP.  You can also tell a story where this could happen through the private sector provided as favorable borrowing terms as (or the ability to set interest rates like) the government.  However, one should be reminded that the likes of Paul Krugman only advocate the building of FEMA death camps when the  market has failed to produce full employment.  In which case the government is also making savers better off because it is giving individuals something to hold  that pays at least some kind of risk adjusted return in olight of a lack of investment.  Individual svaes can be made better off even though on aggregate the whole point is to finance today with something that will be insignificant tomorrow. 


Now, we can debate whether the relevant variables that determine the shrinking of the debt hold or that the inflationary bias is too much, or the size of the multiplier but the basic principle stands.


*For the uninitiated:  GDP is the purchase of newly produced final goods and services.  This is distinct from intermediate goods which are inputs into the production of final goods and services.  The reason for this distinction is becuase final goods and services already contain all the costs of the inputs (and factors of production**) that went into making them.  We use final goods and services as a measure of output becuase final goods and services contain the value of everything that goes into making them and so we get a full measure of the "output" of the economy.

**A "factor of production" is something that does not get "used up" in the production of a good.  This is distinct from an "input" which does get used up.  For instance when a baker bakes a loaf of bread the wheat he uses is an input but his labor (and the oven, his capital) do not get used up are factors of production.  The price of a loaf of bread (a final good if I buy it) contains in it the cost of the wheat, the cost of the bakers labor, the interest he has to pay on the oven he bought with a loan and the profit the baker gets from owning the bakery (all payments to the factors of production).

.*** Presumably though scale inefficiencies are irreverent from a utility point of view since I must be on net gaining at least some utility from going through the trouble to bake my own bread (assuming I have the income to do it)  instead of just buying it.

The Devil is in the Deflator.

At Econospeak Barkley Rosser brings up WWII:

This curiously relates to a matter that has been much argued about previously on various blogs, namely the nature of the short and sharp post-WW-I recession of 1919-20. Some have argued that this shows how wrong Keynes was, because laissez-faire was followed, including letting prices (and some wages) fall sharply in 1921, with the economy bouncing back very nicely, after having the unemployment rate soar from 5% in 1920 to 9% in 1921. Most economic historians have attributed this recession to "postwar adjustment problems."

OTOH, some of those making a big fuss about that recession somehow fail to notice that in fact there was a post-WW-II recession, if also very brief, if sharp. It occurred in 1945 with the sharpest decline in wartime spending, although not much remembered. However, the official stats have US declining in GDP by a whopping -12.7% in that year, although that number must be taken with some grains of salt due to all kinds of measurement issues and restructurings. Some say this exaggerates things as the unemployment rate only went from 1.5% to about 3.6%, a rise, but not all that much to get worked up about.

Two points. The first is that this latter event does not account for the massive decline in female labor force participation that occurred in 1945, from about 38% to about 30%. We all know (or should) that those withdrawing from the labor force do not count in the unemployment rate. That not very large increase in the UR does not disprove that there was a sharp (if short) decline in GDP.  The second point I'm not going to deal with, it is about the Federal Reserve and interest rates after the war. The point above though I'm more than happy to write about. While I do believe Prof. Rosser is mistaken, those that are challenging him in his comments section are right for the wrong reason.

Robert Higgs was mentioned about a dozen times as providing the basis for the idea that there was no post war recession. However, that claim is somewhat dubious. Higgs starts from the presumption that all government spending is useless and wasteful and with that eliminates the decline in government spending from consideration when he declares that there was no post war recession. Essentially Higgs' argument revolves around assuming that "legitimate" or "welfare enhancing" GDP = C + I + NX. What Higgs focuses on is the extraordinary jump in investment in 1946 (the end of war spending shows up in BEA data mostly in 1946, not 1945) which is actually quite remarkable and which Higgs attributes to the "pro-businesst" policies of the new Truman Administration and the general go-gettingness of the supply side.

The conclusion you reach about the depth of the recession when using NIPA data--as Higgs does--depends very heavily on which deflator you choose to use. A cottage industry has sprung up around producing WWII deflators:

Table 1 shows real GDP and real GDP growth given a bunch of different deflators:

Table 2 shows real "private" GDP and real private gdp growth

Starting from the "Hayekian" assumption that there was no business cycle downturn after the war does not disprove that there was no Keynesian effect at work. One of Prof. Higgs' most important insights is that savings was not "spent down" after the war as the traditional Keyensian story tells and that is much of his basis for declaring a supply side explanation to the post war boom . However, the demand side story holds up if one thinks of that pool of savings as the basis for the financing that allowed a returning G.I. an his new bride Rosie the Riveter to fill up there new house with the durable goods that the war spending had crowded out. Much more importantly, that pool of savings provided those new families with the financing to buy the house they bought and needed to fill with modern conveniences.

Another way of putting this that is more in line with the point Rosser is making is that Keynesian theory does not dictated an endless stream of government spending to prop up the economy. One can think of the post-war boom as a delayed multiplier effect. War production crowded out the production of durable goods (most importantly housing and cars) but increased incomes (though, yes, nominal disposable incomes did "stagnate" from 1942-1945 after a 30% increase between 1939 and 1942). These increased incomes became savings without the normal platter of goods available. That savings became goods after the war when the swelled balance sheets of financial institutions looked for new assets to replace wartime assets.

In fact, WWII as a "Keynesian Event" is unique in how dramatically it parses out the effect of government spending increases and then the secondary multiplier effect that provides the autonomous engine of growth that Keynes clearly saw as being at the heart of the capitalist system, the debate over the preface to German edition of _The General Theory_ aside.

The Demand For Treasuries.

There have been a number of things lately such as the SS debate, reading both the General Theory and Hyman Minsky's "Stabilizing and Unstable Economy", and--not least of all--my dissertation that have got me thinking about the demand for treasuries. We are used to thinking about government debt as a supply side phenomenon, simply politicians turning on the printing presses so to speak.

To illustrate, here is a question from my final (both Marco and Money and Banking) that I asked:

In the US right now corporate investment is extremely low as corporations are now holding historically high levels of liquid assets instead of spending their profits. As well, foreign savings continues to flow into the US because US assets/investments increasingly looks good compared to European assets. Finally, private savings rates are increasing as American households try to "repair their balance sheets" and make up for wealth lost when the housing bubble popped. Use the “Savings/Investment Identity” to show what must happen if capital inflows and households savings are increasing while at the same time investment is decreasing.

To answer this question you need this identity:

Investment (I) = Private Savings (PS) + Capital Inflows (KI) +Government Savings (GS)

Necessarily, if the events I outlined are true, Government Savings has to fall, that is, the government has to run a deficit. Now, this is an idenity and it doesn't tell you anything about what is driving what. But the way I framed the identity it is all about the demand side for treasuries. In a weird way, the extension of the Bush tax cuts following on the heals of the Eurozone crisis can be seen as a kind of market response to demand.

Anyway, I decided to dig around in the Fed's Flow of Funds and the Federal Government's Budget data to see what the demand side has looked like historically.

Right away I'm gonna deal with the mot controversial part of this little exercise. I am including the Social Securty Trust Fund as part of the demand for debt. Here is my version of a graph I'm sure you've seen a million times:

Figure 1

So Social Security is about half of the debt that the federal government owes itself. The reason I have included social security is because even though it is classified as debt that the government owes itself it also represents (potentially) future public debt. Social Security debt is held in special "non-marketable" treasuries and so currently they aren't part of the market for treasury bonds and to that extent the debt is actually different than debt to the public.

However, as the Social Security system starts to take in less money than gets paid into it the Social Security Administration will cash in those bonds. In order to pay back those bonds the federal government will have to pay that back with taxes or issue new debt issued to the open market. Social Security is "postponed debt to the public". I need to be careful and point out that this is not a problem with SS. I'm not much of a deficit/debt hawk and to the extent the debt is worrisome the issue is how we manage the debt/GDP ratio. The shift of trust fund debt to the public would have no effect on the debt/GDP ratio. Depending on how rapidly the trust fund got used up it may have an effect on treasury bond yields but that seems to me like mostly a secondary concern.

I don't want to dwell on Social Security. The dudes (gender neutral!) over at Planet Money have a really good podcast about social security. Maybe my next post will be about Social Security. Anyway, I do need to admit a couple of things. First, it's a little unfair to single out social security for my analysis but I had to make my graphs intelligible. The social security trust fund made up about $2.6 trillion of the $4.3 in debt the government owes itself at the end of 2009. So I'm ignoring $1.7 trillion which is a lot even when talking about an $11.8 trillion deficit.

That $1.7 is mostly made up of the:

Civil Service Retirement Fund ($750 billion)
Department of Defense retirement and Medicare funds ($144+ $295 billion)
The Medicare trust fund ($390 billion)

Anyway, the vast majority of the debt the government owes itself is the investment of retirement entitlements of one kind or another. But like I said, i had to keep my graphs reasonably clear so I'm ignoring $1.7 trillion!

Alright so here is the first graph:

Figure 2

This graph is included for completeness but its kind of hard to read because the Great Recession kind of explodes the graph. It's a good place though to look to explain a few things. First, of all these are "flows" which means what is being shown here are either new (net) purchases or sales of treasuries by each sector for each year. Secondly, the "Remainder" line represents the total new treasury issued in each year minus the sectors show in the graph. It just shows how much of the new treasury debt issued was bought by the sectors I'm looking at. If the Remainder is positive it means the sectors shown under account for new treasury issues in each year. If the Remainder is negative the sectors shown over account for new treasuries. Finally, I chose 1975 as the starting point because that's around when the post-war debt/GDP ratio bottomed out.

I'm going to chop off the Great Recession (and come back to it later. Here we have a picture of what should probably be called the structural deficit that started in the early 80s:

Figure 3

In figure 3 I have lopped off 2007-2010 or, roughly speaking, the Great Recession. I have also put everything into 2005 prices. I am not so sure that was a good idea but I felt like keeping it in nominal dollars distorted the relative importance of each sector over time. I would have preferred to make this a logarithmic chart, but I can't because there are negative values. Anyway, this this is essentially unreadable and won't make sense until after looking at figure 4. However, two things are pretty clear. First, the financial industry seems pretty important to the demand side through the 80s. Second, while state and local government (includes "regular budget" and pension funds) sell off a lot of treasuries in the 90s their purchases seem to recover in the mid 2000s. However, in the 90s households start dumping treasuries and don't look back (until 2008!). Did all the money wind up in the stock market? Did treasuries become a quaint, old timey way to save?

Okay so Figure 4 can help make sense of Figure 3:

Figure 4

So figure 4 is not really "to scale" in any way. Each bar represents 100% of treasury purchases by the sectors I'm looking at (plus the remainder) for each year. However, this does not give any sense of the relative scale of each year. For that look at figures 1 and figures 2.

What's interesting about this chart is that there seems to be a clear shift in the demand for treasury bonds that takes place in the mid 90s. It looks to me like "core demand" shifts from the financial industry to foreign demand. As well in the 90s there seems to be shift of demand from state and local government to Social Security. However, remember these are all shares of a pie that is growing every year. If you look back at figure 2 it seems more that state and local demand remains steady (though falls a little) while SS demand starts to take off in the late 80 due to the 1983 Social Security Amendment.

I found the results from figure 5 to be pretty surprising:

Figure 5

I chose 1971 because that's when the US went off the gold standard (officially) and I was curious to see if that meant anything. Now what'ss obvious here is how important foreign "official" institutions (mostly foreign central banks) are as a source of treasury demand. I had taken the role of the US dollar as a "reserve currency"--as a currency other central banks hold as a way of managing their own currencies-- for granted. However, it is quite striking how important our role as a reserve currency is in creating demand for our treasury bonds. Other countries prefer to hold treasury bonds instead of dollars because they are essentially as safe as dollars and they pay a return.


I want to go over two things. First, I don't think this post would be complete without a graph of the last couple of years (at least through 2009, which is all I have numbers for):

Figure 5

The scale of figure 5 is nominal dollars. The Fed sold off a whole bunch of treasury bonds in 2008 in order to purchase (or sometimes exchange) for things that were not treasury bonds like a Mall in Oklahoma (Planet Money Again). Foreign demand went through the roof and financial business came back to buying treasuries. Foreign demand is probably going to get stronger now that Europe is becoming more and more of a mess, a phenomenon called "flight to quality". As far as the financial industry buying treasuries that's a function of the Fed loaning tons of money to banks basically for free. Banks and other financial intermediaries then take the money and buy safe as houses treasury bonds and take home the difference. It helps keep treasury bond yields low and Wall Street bonuses flowing.

Finally, I'm not 100% sure I've done a good job making it clear how important governments are to the demand for treasuries. Here is one more homemade chart:

Figure 6

I double checked this graph after I made it because I find it really surprising. I feel that this really makes it clear how important political entities are to the demand side of the treasury debt market. I know I'm trying to play both sides of the fence here by saying that SS should be thought of as simply postponed debt to the public and then putting it up here as part of "political" debt. But I think its justifiable because a lot of the discussion these days is about "the market" for treasury bonds and how market bond vigilantes may soon send us the way of Greece and Ireland. However, the vast majority of treasury debt is dependent on political decisions and political entities (in 2009, the gray area of figure 6 was only $3.4 trillion of 11.9 trillion). Here again Social Security is useful as illustration. While I do believe Social Security is "real debt" the political debate today seems to be all about trying to avoid having to turn these non-marketable securities into "general fund" debt very quickly or at all either by cutting benefits or by raising current taxes. As well, it seems highly unlikely that the Chinese are going to stop pegging the Yuan to the dollar any time soon even if they are going to tweak it here and there. As well, the Fed is obviously extremely sensitive to what it does to the treasury market and even a good portion of the private market debt are essentially back door "open market purchases" made through arbitraging financial intermediaries.

Now, I'm not willing necessarily to call demand based on political decisions more stable, but it's not obvious that even if the fantasy of bond vigilantism came true that they would have necessarily that much of an effect.

Oh and also on the "we aren't Greece or Ireland" tip. I feel like this is a good place to put this graph from the Financial Times:

Figure 7

The basic point to be made here is that the analogy is not between the US and the second tier developed nations of the Eurozone. The proper comparison is the comparison between the US, the financial center of the world with the world's reserve currency is with the UK when it held that position. However, I do have to admit it's not entirely clear which UK we are talking about. The post Napoleon UK was an empire experiencing very robust growth on the whole. The post Hitler UK on the other hand was the shell of a collapsed empire that eventually needed to be bailed out by the IMF in the mid 70s.

it should be pointed out that the post WWII UK had lost it's prominent position as the worlds financial center with the world's reserve currency. The US remains strong on both fronts. While there is much talk of China, their financial markets are still in their infancy and you aren't even allowed to trade Yuan. Before their own subprime crises the EU looked like it could possibly be a contender but that seems far more remote now although this crisis may bring the Eurozone closer to a fiscal union which they would need to compete effectively with treasury bonds. At any rate the US seems pretty secure as the worlds tallest midget.

Why QE2 and reserves dont matter.

Auerback: Amateur Hour at the Federal Reserve

By Marshall Auerback, a portfolio strategist and Roosevelt Institute Fellow

As any student of Economics 101 realises, you can control the price of something, or the quantity, but not both simultaneously. In announcing its decision to purchase an additional $600bn of treasuries last week, the Federal Reserve presumably intended to create additional stimulus to an economy, since tepid growth has failed to make a dent in unemployment. Even Friday’s “good” unemployment numbers, where the US economy added 151,000 jobs, were not enough to reduce the current jobless rate of 9.6%.

So is a new round of “QE2” going to do the trick? It would be interesting to figure out how the Fed came to the magic number of $600 billion. Why not a trillion? Why not $250bn? Why $75bn a month? There’s an element of sticking one’s finger in the air and hoping for the best. The Bernanke Fed is slowly reaching Greenspan-like levels of incompetence.

Let’s go back to first principles: Quantitative easing involves the central bank buying financial assets from the private sector – government bonds and maybe high quality corporate debt. In this particular instance, the Fed has announced it will buy $75bn of treasuries a month. So what the central bank is doing is swapping financial assets with the banks – they sell their financial assets and receive back in return extra reserve balances. So the central bank is buying one type of financial asset (private holdings of bonds, company paper) and exchanging it for another (reserve balances at the central bank). The net financial assets in the private sector are in fact unchanged although the portfolio composition of those assets is altered (maturity substitution) which changes yields and returns.

Central bank demand for “long maturity” assets held in the private sector reduces interest rates at the longer end of the yield curve. These are traditionally thought of as the investment rates. This might increase aggregate demand given the cost of investment funds are likely to drop. But on the other hand, the low rates reduce the interest-income of savers who will reduce consumption (demand) accordingly.

Essentially, then, you have a supply side response to a problem of aggregate demand. The cost of investment funds might well drop, although that’s not 100 percent clear. Consider the following example: Let’s say the Fed simply targeted the 10 year treasury at 2.25%. The central bank would have a bid at that level and buy all the securities the market didn’t want to buy at that level. They may in fact buy a lot or a very few, and possibly none at all, depending on Treasury issuance, investor demand, and market expectations.

But the FOMC has announced a limit to the Fed purchase program, both in terms of the monthly amounts and the total quantity. How high will 10 year notes trade with the billions free to trade at market levels?

It could be at much higher yields, which presumably defeats the whole purpose of the program. Of course, if the Fed bought every single 10 year treasury, then for sure they could maintain that rate indefinitely; but then they would have to preclude announcing a specific amount that they wished to purchase. Of course, if the Fed did this, people would undoubtedly squawk about “printing money” and “creating inflation” but again, QE does not actually create new net financial assets.

In fact, the Fed has done nothing but TALK about its plans over the past several months, but has yet to initiate the program. There has been no widespread “money printing” or “currency debasement”. The Federal Reserve announced an INTENTION to do something and private portfolio investors took their cue from that. But as any Asian central banker can tell you, private portfolio preference shifts are notoriously fickle, with conflicting motives. The term structure of rates in Japan would imply a comparable intent to make holding yen unattractive, yet 10 year government bonds in Japan yield less than one percent and the yen remains resolutely strong against the dollar.

Is there another method by which the Fed could influence the long term structure of interest rates? Why not just stop issuing 10 year government paper? If the Treasury had announced they were eliminating everything longer than 2 year notes for new issues there would hardly be any screams of “money printing”?

And how low would the 10 year go?

If the objective is to allow the banks to earn their way out of insolvency, then QE2 is also an ineffective means of doing this, since it flattens the yield curve, but still engenders interest rate risk on the part of the banks which persist in “playing the yield curve”, especially via leverage. There is a more effective means of doing promoting bank profitability in a comparatively risk free way, as my colleague, Randy Wray, has noted:

[T]here is nothing wrong with offering longer-maturity CDs to replace overnight reserve deposits held by banks at the Fed. Banks are content to hold deposits at the Fed—safe assets that earn a little interest. They are hoping to play the yield curve to get some positive earnings in order to rebuild capital. If they can issue liabilities at an even lower interest rate so that earnings on deposits at the Fed cover interest and other costs of financing their positions in assets, this strategy might work. That is what they did in the early 1990s, allowing banks that were insolvent to work their way back to profitability. The Fed could even lend to banks at 25 basis points (0.25% interest) so that they could buy the CDs, then pay them, say, 100 or 200 basis points (1% or 2% interest) on their longer maturity CDs. The net interest earned could tide them over until it becomes appropriate for them to resume lending to households and firms.

To be clear, we are not necessarily advocating banks play the yield curve to restore profitability (far better to have a payroll tax cut for that), or that they get such an arbitrage from the Fed, only that it would work with less risk both to the bank and the financial system than what will result from QE. Ultimately, if the objective is to allow banks to restore profitability via traditional lending activities, then there are far more obvious ways to do so. Let’s first recall that BANKS DO NOT LEND THEIR RESERVES, as is always depicted in the economics textbooks via so-called “fractional reserve lending”. The Federal Reserve Bank is a bank, just like Citibank or Wells Fargo. The Treasury has an account at the bank, just as you or I have a checking account at our local bank. Other banks, like BofA and Wells Fargo have accounts at the Fed as well, and they are called reserve accounts (for more see here).

Reserve accounts are not made up of money held in reserve in case a loan goes bad, they are money held at the Federal Reserve for payment settlement. The reserves of money held in case loans go bad are capital. They are not lent out. The way banks actually operate is to seek to attract credit-worthy customers to which they can loan funds to and thereby make profit. What constitutes credit-worthiness varies over the business cycle and so lending standards become more lax at boom times as banks chase market share. But that’s a function of credit analysis (or the lack of it, as the cycle matures), NOT the bank’s reserve positions.

These loans are made independent of their reserve positions. At the individual bank level, certainly the “price of reserves” will play some role in the credit department’s decision to loan funds. But the reserve position per se will not matter. So as long as the margin between the return on the loan and the bank’s marginal cost of funds is sufficient, the bank will lend

So as long as the margin between the return on the loan and the rate they would have to borrow from the central bank through the discount window is sufficient, the bank will lend (for more explanation, see here).

Consequently, the idea that reserve balances are required initially to “finance” bank balance sheet expansion via rising excess reserves is inapplicable. A bank’s ability to expand its balance sheet is ultimately dictated by the presence of creditworthy borrowers, which is a function of FISCAL POLICY. Instead of expending political capital on pointless accounting shuffles and financial guarantees, or programs such as TALF, TARP, Term Auction Credit, or the Commercial Paper Lending Facility, one could simply implement a $2 trillion tax cut (or spend it via the government, or do revenue sharing with the states), thereby generating increased spending power in the economy, providing significantly higher multiplier effects and ultimately shifting the economy back to self-sustaining growth. This would be far more effective than continuing an endless array of silly Fed programs, which do nothing but diminish the central bank’s credibility and have had minimal impact in terms of economic growth.

In many respects, the markets are already implicitly conceding the ineffectiveness of “QE2”. Within hours of the program being announced, there were stories of “QE3”. Instead of having a situation where the markets “buy the rumor, sell on the news”, it appears to have been more a case of buy the rumor and then double up on the news.

We’ve truly hit amateur hour at the Fed. We’re inciting speculation and the Federal Reserve is acting like the kid in his car seat who keeps turning his toy steering wheel as much as it takes to turn the car. Toy cars, however, won’t get you very far if you plan a long journey, and likewise QE2 is a pretty ineffective vehicle if one wishes to engender genuine economic growth. Eventually, investors will realize they’ve been conned (yet again) by the Fed and the end result won’t be pretty.

Don’t Fear the Rise in the Fed’s Reserve Balances

By Scott Fullwiler

Many in the financial press have noted the rise since September 2008 in the Fed's reserve balances from about $20 billion to more than $800 billion today. A number of well-known economists have expressed concern that this will be inflationary.

However, fears that these are inflationary are misplaced, even inapplicable, as they apply only to a monetary system operating under a gold standard, currency board, or similar arrangement, not the flexible exchange rate system of the U. S.

Under a gold standard, for instance, banks must be careful when creating loans that they have sufficient gold or central bank reserves to meet depositor outflows or legal reserve requirements. This is the fractional banking, money multiplier system standard in the economics textbooks. If there is an inflow of gold, then bank deposit creation can increase and prices can rise. The same can occur if the central bank raises the quantity of reserves circulating relative to its own gold reserves.

But that's not the case under modern monetary systems with flexible exchange rates.

In the U. S., when a bank makes a loan, this loan creates a deposit for the borrower. If the bank then ends up with a reserve requirement that it cannot meet by borrowing from other banks, it receives an overdraft at the Fed automatically (at the Fed's stated penalty rate), which the bank then clears by borrowing from other banks or by posting collateral for an overnight loan from the Fed. Similarly, if the borrower withdraws the deposit to make a purchase and the bank does not have sufficient reserve balances to cover the withdrawal, the Fed provides an overdraft automatically, which again the bank then clears either by borrowing from other banks or by posting collateral for an overnight loan from the Fed.

The point of all this is that the bank clearly does not have to be holding prior reserve balances before it creates a loan. In fact, the bank's ability to create a new loan and along with it a new deposit has NOTHING to do with how many or how few reserve balances it is holding.

In other words, there is no loan officer at any bank that checks with the bank's liquidity officer to see if the bank has reserves before it makes a loan.

What constrains a bank in the creation of new loans and deposits, then? First, there is the fact that there must be a willing borrower . . . one whom the bank deems to be creditworthy. Second, the loan must be perceived as profitable . . . in this case, the bank's ability to raise deposits does matter, since it probably expects the borrower to withdraw the deposit it will create, and finding new deposits is much cheaper for the bank than borrowing from other banks or from the Fed. Third, the loan must be on the regulator's approved list of assets, and if the loan results in an expansion of the bank's balance sheet, the bank must be aware of the impact on its capital requirements and other financial ratios with which the regulator is concerned.

But, how many reserve balances the bank is holding does NOT affect its operational ability to make the loan.

Most fears expressed by economists, policymakers, and the financial press regarding the rise in reserve balances since September presume—like the inapplicable money multiplier model—this will necessarily lead to excessive creation of loans and deposits by banks and thus rising inflation.

But this cannot possibly be true. Banks have the same ability to create loans with $800 billion in reserve balances that they had with $20 billion. The difference now is mostly that they do not see as many creditworthy borrowers coming through their doors, given the deep recession, which has led them to create fewer loans.

Admonishments of banks by members of Congress for "not lending out the TARP funds" make the same mistake. Banks don't lend out TARP funds or any other funds. They create loans and deposits out of thin air, then use reserve balances to settle payments or meet reserve requirements.

For further evidence, consider two recent extreme cases:

In Canada, reserve balances have been effectively zero for over a decade now, and bank lending continues as it does anywhere else. Canada's inflation also has been similar to that of the U. S.

In Japan, under the so-called quantitative easing regime of 2001-2005, reserve balances reached around 15% of GDP, and the monetary base (reserve balances plus currency in circulation . . . often termed "high powered money") reached 23% of GDP. But Japan has, if anything, experienced deflation during and since this period, which is not surprising, since—again—the rising quantity of reserve balances did not enhance Japanese banks' abilities to create loans.

In the U. S., by comparison, reserve balances have reached about 6% of GDP, with the monetary base rising from about 6% to about 12% of GDP since September 2008. Those fearing rising Fed reserve balances apparently haven't noticed that an increase in reserve balances about three times the size in terms of GDP already happened in Japan, with none of the effects that have been predicted for the U. S.

In short, don't fear the rise in the Fed's reserve balances. It is not inflationary because the money multiplier view, found in the textbooks, doesn't apply to the flexible exchange rate monetary system of the U. S. The U. S. may indeed experience rising inflation in the future (or it may not), but it won't have anything to do with the quantity of reserve balances banks are holding.

Data sets for dissertation

Fed state level data

http://fraser.stlouisfed.org/publications/allbkstat/

Mason and Calomiris:

http://research.stlouisfed.org/publications/review/98/05/9805jm.pdf

FDIC member bank data:

Brief explanation of housing in NIPA:

http://www.michaelcarliner.com/HE0110-Hous-GDP.pdf