Search This Blog

Wednesday, October 23, 2013

Crowding Out And Its Relation To Bullshit

A topic I've been hearing from some of my conservative friends is a refrain often found in mainstream macroeconomic textbooks - crowding out.  Crowding out is the theoretical idea that there is a fixed pot of gold from which to finance investment, so if the government all of a sudden wants to draw from that pool, it must mean it has to take funds from the private sector (because there's only so much gold to go around).

So a few things are wrong with this idea:
1. We are not on the gold standard anymore.

2. In our economy there are funds that are just lying around because no one is using them.

3. Even if the economy was better and the private sector were not hoarding cash and if they were instead fully utilizing all resources, there is no fixed 'pot of gold'.   The amount of funds in circulation are completely driven by demand.  So, if the government wants to spend more money, they don't 'take it' from the private sector.   They just spend the money and US bonds are issued - with no effect on the amount of resources whatsoever.

But people don't seem to get this.  Especially conservatives.

But as I've been trying to convince people for years, the reason that conservatives don't get it is because mainstream economists don't get it, and they keep teaching bullshit... like crowding out.  I keep teaching bullshit like crowding out too, because I'm virtually forced to - but I also try my best to say in a very professional way, "this is bullshit".

And then the brainiacs at the Congressional Budget Office say things like this (from 2009):

"ARRA's [Obama stimulus] long-run impact on the economy will stem primarily from the resulting increase in government debt. To the extent that people hold their wealth in government securities rather than in a form that can be used to finance private investment, the increased debt tends to reduce the stock of productive private capital."
I have bolded that last sentence for you.  What it is saying is that crowding out will happen, but what it also implies is that crowding out will happen only under certain assumptions: "To the extent that people hold their wealth in government securities rather than in a form...." implies that there is an either/or situation.   The CBO is making this assumption, which I just told you is false - and that every central banker will tell you, is false.   There is not either/or situation between bonds or savings that businesses can use.   Businesses don't finance their investment from some fixed pool of money.   

Can you imagine if a business went to a bank for a loan, made the business case for the loan (which was a good one), but then the bank says, "Oh I'm so sorry sir, but we've plum run out of money.  I mean, we had some, but unfortunately the government sold so many treasury bonds that we now have none left."   I mean - seriously?   Has this ever happened? No, it hasn't.  Do you know why it hasn't?  Because loans are not made from some fixed pot of gold. Loans are made based on risk and return and the business case, and that's it.   Banks never 'run out of money'.   That's what the Federal Reserve does - ensures that banks have the money they need to do business.  

Having said all this, is it possible that in some cases government spending influences business incentives to a degree that might cause less private investment.   Yes, probably.  But, that's not what the mainstream theory says.  The mainstream theory is crowding out occurs because there is a fixed pot of gold so banks won't have resources to make loans, and interest rates will rise so businesses won't want to borrow, and taxes will rise in the future so that the average Joe will hoard his cash that would otherwise go to US businesses.    All of that theory was born at a time when we actually did have a fixed pot of gold.   But today, all of that, is bullshit.

Thursday, October 17, 2013

The True US Debt Burden

In macroeconomics, it is often taught that that the US debt burden (ie., the metric we should gauge how concerned we should be about the national debt) is the amount of interest paid on said debt divided by the measure of GDP during the same time period.   This makes sense.   We can roll over the principal of our debt and we can just use some of our growing economy to 'pay for' the interest we owe and it's 100% sustainable forever with no real cost to anyone.   As long as the growth in the interest we have to pay to the bondholders does not exceed our economic growth (it would have to be decades for it to really matter that much), we are right as rain.

But then, many textbooks, including the one I use to teach my macro class go on to say the best 'rule of thumb' therefore is to try to keep a constant or shrinking debt to GDP ratio - not interest on the debt, just debt.  This I have a big problem with.

The problem with using that as a rule of thumb is that it is not very useful, mainly, because it assumes that interest rates paid on the debt remain constant.  But as we all know, interest rates are anything but constant and in fact real interest rates have drifted negative in the past few years in some cases to the point that the government actually can theoretically earn money by going in to debt.  Even if that is not the case right now, the fact remains that real interest rates have dropped dramatically (thanks to the Fed) in the past 6 years.   This means that any new government debt is dirt cheap.   IE., interest/GDP aka the debt burden cannot be nor should be approximated by using debt/GDP as a measure.   If what you care about is the debt burden, and as long as the government collects such information, why would you need another measure of approximation anyway?   I mean, unless you are trying to scare people:


The CBO, as an aside, believe the debt/GDP ratio will fall for the next decade or so before it starts rising again, but it's also important to note that the rise precludes any assumptions about the investments we are putting in to the economy today.   So, really, predicting this sort of thing a decade out is a fools errand anyway.

But, why scare people, when you can present reality:
The reality is we aren't paying anything more on interest to our debt really now than we were back pre-9-11.   IE., thanks to a recovering economy and falling interest rates, despite our increased debt values, the debt burden is actually falling and really has been since the 1990s if you take a longer-term view largely due to the tech boom, the end of the cold-war, etc., which despite the bust, still has aided our long-term economy to this day:


Say it with my tea-publicans:  the economy drives the debt burden, not the other way around.

Thursday, October 10, 2013

How Classical Economics Misleads On "Government Debt"


The classical model is mandatory reading for most macroeconomics classes.  I teach it because I have to, but I teach it in a way that the caveats and assumptions are front and center.   It's apparent to me that the profession of economics perhaps has not drawn enough attention to those assumptions (or perhaps exacerbated the view that they are unimportant), since it seems a significant chunk of the American public is hyper-concerned about the debt, when perhaps such concern is not warranted compared to all our other problems du jour. 
The classical economic model is a ‘pot of gold’ model, where an economy has a certain fixed amount of resources by which it creates productive capacity.   The classical model assumes markets are perfect in the sense that there is 100% utilization of resources at market determined costs of those resources.  This means that at market given wage and interest rates, we have ‘full employment’ (not counting a certain amount of unemployment that may exist just due normal movements between jobs, etc.) of both labor and physical capital (buildings are staffed, machines are manned, etc.). 

So, now, in this classical world the government comes in to an economy that is already using 100% of the resources and says, “I want to spend $1B more and give everyone healthcare.”   In our classical world, there are no more resources to devote to this new policy so the government has to take it from the existing ‘pot of gold’.  In other words, they have to take resources that are currently being (presumably productively) used in the private sector (by businesses).   In this world, at a minimum, the healthcare spending does nothing for growing the economy and at worst it does one of two things (or some combination):  (1) severely harms economic growth by taking some of the existing ‘gold’ necessary for the private sector to do its thing – either by increasing taxes or some other mechanism, and/or (2) in the case where the government starts up the printing press and starts manually artificially creating more 'gold' to fund its spending, inflation will ensue.   Neither of these is pretty.    In either case, the classical model further assumes that when the government spends money, 0% helps economic growth, while, when businesses spend money, 100% of that spending helps economic growth. 

But, the real world doesn’t look anything like the above classical ‘pot of gold’ scenario.  In the real world, economies are subject to bouts of booms and busts for varying reasons: sometimes there’s an asset bubble that bursts (housing, stocks), sometimes there is an oil crisis that causes prices of all business activity to raise thereby stunting growth, and sometimes there is a large degree of uncertainty and other more ‘soft’ things that create pessimism in markets – reducing the revenues businesses need to operate and grow effectively. 

In the real world, therefore, economies often are not operating at ‘full employment’.  Perhaps a significant chunk of the labor force is out of work.   Perhaps perfectly usable buildings are sitting empty for months.   Perhaps banks aren’t lending and perhaps businesses are just sitting unproductively on their assets.   This seems like common sense reality, but it’s this common sense fact that changes the outcome of the government wanting to “spend $1B and give everyone healthcare.”   Now, the government can enter into a deficit situation not by taking from the private sector but simply by starting up the printing press and creating more funds.  They could do this in the classical world too but unlike the classical world this will not result in inflation because the increased spending/demand from the new government policy simply soaks up some of the already underutilized resources (it results in workers being hired to enact the policy and carry it out, and it results in certain capital being used that might otherwise be sitting idle).   In this kind of environment, there is enough 'supply' of productive capacity to fill the 'demand' for this new government program, and no inflation needs to occur.

But what if this healthcare policy were enacted in a ‘normal’ economic time – not on the tail-end of a recession?   Even if we assume that 'normal' means we have 100% full employment (which is itself debatable), classical economics falls apart.  The key here is the unfortunate assumption classical economics often makes about the difference between when the government spends and when businesses spend.   Recall from above, in the classical model, when the government spends money, 0% helps economic growth, while, when businesses spend money, 100% of that spending helps economic growth.   Most folks recognize that assumption to be patently false in the real world.   When BP invests millions of dollars in off-shore oil rigging that eventually causes multi-billion dollar environmental catastrophes, or when Chinese businesses invest money in making tires for export to the US that subsequent explode while us Americans are barreling down the road – not all of that sounds particularly productive: quite the contrary actually.  Conversely, when the government spends money on educational infrastructure, transportation and logistics infrastructure, or perhaps even providing some certainty for millions of Americans regarding their healthcare…. Some of that is likely very productive and adds to the economic capacity of the United States.   The point here is that some government spending actually increases the size of the pot.   And to the degree that the potential for that growth that might not normally occur in the private market is probable compared to the potential for the policy to cause inflation – even in ‘normal’ economic times that spending policy may be worth it.   In today's environment, the potential of Obamacare to increase economic growth is debatable.   But the potential for it to cause inflation is almost non-existent because (1) it is of relatively small cost in the grand scheme of our economy, and (2) if the markets thought significant inflation was on the horizon, interest rates on non-government backed securities would be pressured to increase as potential lenders would demand a higher rate of return to cover the higher future prices - but this is not happening at all today.
 
What this means for our present political debate over the government debt is that there is a fundamentally simplistic worldview that many Republicans have, that is largely based on the overly simple classical model of economics: the idea that when the government goes more into debt, it takes existing resources from the private sector and kills economic growth.  It ignores the reality that in fact, the debt is much more a by-product of economic growth and the cycles of booms and busts as supposed to the other way around.   And that in fact, sometimes, the government can have positive effects on economic growth albeit with a substantial upfront cost.   Most folks call that an 'investment' in the future.