Thursday, 25 February 2016

Why it is so important to use government deficits rather than negative rates to boost the economy (demonstrated with a lot of pretty pictures).

Many economists treat fiscal and monetary policy as equivalent, to be used interchangeably. This article by Martin Sandbu in the FT is an example, and he would have the support of a majority of academic economists in this view.

But it is wrong; they are not in any way interchangeable. The reason the majority of academic economists are wrong is that their models for the economy don't include debt. Because their models do not include debt, they are unable to explain why we are at zero interest rates (too much private sector debt), why we have secular stagnation (too much private sector debt), growing inequality (partly too much private sector debt), too high house prices (too much private sector debt) and I could go on, but I imagine you get the point.

To remedy this, I have created a very simple model, called a Demand Based Cash Flow Model (DBCF). It is much simpler than the DSGE models economists normally use. It has one major advantage, though, in that it does include debt.  It is based on simply looking at cash flows on the demand side of the economy, with an adjustment made to account for supply issues. And, in my opinion, it is a very powerful forecasting tool (relatively speaking, of course; the bar is set pretty low in economics).

Recently I was asked by Professor Robert Hockett at Cornell, if I could use the model to examine the spending plans of the US presidential candidate Bernie Sanders. In doing so, I programmed the model into a spreadsheet and found some very interesting results, which I will share below.

I will run three different scenarios for the US economy. In the first one, interest rates remain at the zero lower bound, and government deficits are kept at approximately the current level. The second will use monetary policy to stimulate growth. The third will use fiscal policy.

The spreadsheet is available on request if anyone wishes to see the assumptions made. In general these are taken from the DBCF model paper, which I recommend reading if you are interested in how these results were reached.

SCENARIO 1: NEUTRAL. The government runs a deficit of 4% of GDP, interest rates remain at zero.

In this scenario, we continue in the secular stagnation we have been in. Economic growth continues at around 1.8%:

Inflation recovers from the current lows, which are largely due to deflation in commodity prices. These declining prices cannot continue forever, however, and sticky wages mean that a low level of inflation can survive in very unfertile economic soils. Inflation does remain below the 2% target that the central banks have set themselves and growth can only be described as anaemic.

Average growth in this stagnant scenario is 1.8%. Central banks would be under pressure to try unconventional monetary policy.

SCENARIO 2: MONETARY POLICY. Same as Scenario 1, but cutting interest rates by 0.25% every year.

In this scenario, we imagine that interest rates can go as negative as required and that the effect is the same as reducing rates when above the zero bound. This is debatable, but not completely unreasonable. The lower interest rates work in three ways. First, they stimulate more private sector debt, which puts a flow of new money into the economy. Second, they increase asset prices, meaning more of a wealth effect and more savings that can be spent by people with assets. Third, they reduce the exchange rate for the currency which improves competitiveness. All of this is modelled in the spreadsheet.

This scenario, at first glance looks better (albeit at the end interest rates are approaching the negative 3% level):

Growth ticks along at around 2.5 %, declining slowly but not so much that anyone would notice.
And inflation?

That too is looking pretty good - exactly on target at 2%.

THE GREAT MODERATION! Rejoice all as the economy has officially been solved. If only we can keep reducing interest rates by a further 0.25% every year then we will have endless stable growth.

Of course, house prices have gone up even more and inequality is rising, but maybe those affected should just work harder. We have a great economy.

But not so fast...

This is the graph of private and government sector debt to GDP:

Private sector debt is back up to the levels of 2008. What if we have a crash?

Eventually if debt keeps building up, we must have a crash. Sorry, but this is true; debt can not grow forever, and, as we all know, what cannot go on forever must stop.

Let's put a crash in, in 2021:


Well, that needs some government spending to bail out the banks and keep the economy going. Let's put an extra 10% government deficit in 2021, 7% in 2022, 5% in 2023 and 2% in 2024 (this is on top of the neutral 4%).

We now get a growth chart that looks like this (actually similar to 2008):

With a debt to GDP chart that looks like this.

We have now had an average growth over the period of 1.7% and at the end are left with private sector debt to GDP of 230% and government debt to GDP of140%.

This is not pretty.

So… what if we had used fiscal policy?

Scenario 3: FISCAL POLICY. Run total average deficits of 7% per year for first 6 years and 6% thereafter. Raise interest rates by 0.5% per year

To clarify here, in order to get a good economic impact from running deficits, it is important that the money goes to people with a high marginal propensity to consume. Tax cuts to the very rich will create a deficit but not the demand required. Assuming from hereonin that the deficits are used to fund spending with a high multiplier (assumed to be 1 on NGDP)

In this scenario we get positively healthy growth:

Inflation is a little over target, but not damagingly so:

But importantly, look at the debt to GDP ratios:

Private sector debt to GDP is down from 195% at the beginning to just 121% now. Wages will be a higher share of GDP, meaning real wages will rise. House prices will be lower relative to GDP so those without assets will be able to own their own homes. Overall the economy will be healthier.

And despite all of the extra spending government debt to GDP has hardly moved; it started at 120% and finished at 120% of GDP. The reason for this is because debt is divided by GDP, and GDP has gone up far more than in the other scenarios.

The average real GDP return over the period is 3.5% as opposed to 1.7% in Scenario 2.
The total real GDP growth of the two strategies are shown in the below graph:


I hope this helps to demonstrate why austerity is such a dumb idea.


  1. The entity concept in accounting matters.

    What does the private parties exchange for currency?

    What does the goverment receive for deficit (currency created and spent)?

    This can be analysed using double entry bookkeeping and the fundimental entity concept in accounting. Each entity has seperate transactions recorded in their own seperate journal and ledger.

    "One of the most basic concepts used in the preparation of financial statements is that of the accounting entity."

    Answer 1, Goods and services (time, talant, and knowledge.)

    Answer 2, The same.

    In economics this is called crowding out. But, most people consider this in other realistic terms.

  2. So, the argument in the title addresses liquidity. But, most arguments on the internet do not address the cost of what is traded for that liquid currency or the deficits.

  3. "Andrew Carnegie, Industrial Philanthropist", p. 21, third paragraph.

    p. 32 explains where he was taught to be a capitalist. He was encouraged to buy corporate stock with a loan from his boss.

  4. I wonder if you will make a blog post looking at Japan as it has been doing Keynesian stimulus with a high ratio of private sector debt to GDP.

    1. Japan is the same as everywhere else. They have just been running with very high levels of both private and public sector debt so their saving rate is very high. They need to run very large government deficits.

      Althought they are running relatively high government deficits, they actually just aren't big enough to combat the saving rate. They are always looking to increase VAT or cut spending at any sign of economic upturn. And they are using monetary policy too much rather than fiscal policy.

      In fact with Japan's government debt being so high, any spending only needs a multipler of around 0.45 to reduce the govt debt to GDP ratio. Targeted spending should have a multiplier of well over 1. As everywhere the political fear of larger deficits has been a hindrance.


Sorry, I have had to moderate comments because of an annoying pest spammer who keeps posting American football matches links.