Direct Monetary Financing, Quantitative Easing and Inflation

[Maphylius]

As a political economist, I am very glad that monetary policy has finally been added to the game, even if only in a very limited capacity, which is understandable. The most notable limitation, which has been recognised by the developers, is that there is no central bank target interest rate that can be manipulated. I have some suggestions on how to make monetary policy more realistic with very few changes to the current system and without requiring a central bank target interest rate.

Direct Monetary Financing

Add an additional monetary policy option besides Quantitative Easing (QE) and Helicopter Money (HM): Direct Monetary Financing (DMF). DMF means the newly created money is not pumped into asset markets (QE) or given to citizens (HM), but is used to finance public expenditure. In the game this should be reflected in the budget and influence inflation. One could argue about whether it should also boost GDP, but I would suggest DMF does not do so directly. It can instead do so only indirectly, through the policies that can be financed with it.

It is odd that it is currently possible to hand out money to citizens, but not to use it for public spending.

Quantitative Easing and interest on government debt

QE and similar programs of asset purchases have the effect of lowering the interest rates governments pay on their debt, which is currently not modelled in the game. This is because these programs often include buying of government bonds, which pushes up their price and thus lowers yields, i.e. interest paid by the government. This should be simple enough to fix: just make it so that QE reduces interest rate paid on debt. I think doing this would also to some extent make up for the lack of a central bank target interest rate (for reasons I won’t go into, basically central banks regulate real interest rates towards target interest rate also through asset purchases and sales, so it’s not dissimilar in terms of effect).

Inflation

The developers have recognised that central banks target some inflation (usually around 2%) because too low inflation or even deflation is bad (people hold onto their money instead of spending it). So when inflation is low, expansionary monetary policy (QE, HM, DMF) can be used to counteract this.

What is, as far as I know, currently missing from the game, is that this is often specifically necessary during economic downturns, which can lead to deflation, in the worst case a deflationary death spiral, unless acted against using expansionary monetary policy. I thus suggest that inflation is somehow linked to the business cycle. For example it could be linked to employment, as suggested by the Philips curve, i.e. low employment -> deflation, high employment -> inflation. Perhaps this is already the case, then just ignore this last point.

[cliffski]

its true that we do not model deflation, or have that many connections to inflation either, although we have to keep in mind that many players of the game are quite casual players, who probably understand how inflation works, but have likely never experienced deflation and would not be immediately aware of the implications.
Balancing realism against playability is a real challenge when designing this game

I’m interested in your comments about QE reducing the interest on government debt. Is this really seen to be true? if anything I assumed the opposite. Obviously its increasing the money supply to some extent, but because that is done in a way that specifically offsets bond purchases, it seems to me that the principle impact is a stock market bubble.
Surely the interest charged on government debt is determined primarily by the likelihood of default and the availability of other safe investments? I thought that a government that is using QE is effectively deliberately devaluing its own currency, which would make the debt less valuable, and thus push interest on that debt up?

If governments could reliably reduce interest on their debt with QE, then QE has virtually no downsides and can continue indefinitely, but that cannot be true.

[Maphylius]

Thanks, for the response. I absolutely understand the challenge of balancing realism vs playability, which is why I have tried to make relatively simple suggestions.

I don’t think you need to include deflation in the game at all. My point was simply that economic downturns often lead to deflation or at least lower inflation and this allows governments and central banks to adopt policies which would otherwise be too inflationary. Printing lots of money may be a problem if inflation is already at 10%. But if it is around 0%, then printing money may actually be beneficial to reaching a target rate of 2%.

So my suggestion is that the inflation rate should be lowered during economic downturns, so that players can counteract the downturn with expansionary monetary policy. That’s exactly what is done in the real world and would add an interesting component to the game. The changes required for this would be relatively simple as well. All you have to do is to inversely link the inflation rate to unemployment. This is suggested by the widely-accepted Philips Curve. You don’t have to include deflation at all. 0% can be a lower limit for inflation for simplicity, so that you don’t have to model the effects of deflation.

Quantitative Easing reduces interest on government debt

You have to consider how bonds work. When the government issues a bond in 2021, the bond might say something like “the holder of this bond will receive £110 in the year 2022”. The government then auctions the bond to get the highest possible price, let’s say that is £100. The yeild/interest rate derives from the difference between the money given to the government in 2021 and the money paid back by the government in 2022. So this would give us an interest rate of 10%.

But when the Bank of England prints money to buy bonds (QE), it increases demand for those bonds and thus pushes up the price of the bond. That’s a simple matter of the law of supply and demand. So instead of selling the bond for £100, the government might now be able to get £105 for it in 2021. But it still only has to pay back £110 in 2022. Thus the yield/interest rate is lowered. The Bank of England recognises this in their explaination of QE: “Large-scale purchases of government bonds lower the interest rates or ‘yields’ on those bonds.”

You make the good point that if that’s true it almost seems like there is no downside to QE. And I would say that QE is very powerful indeed. That’s why central banks around the world are using it as much as they are. But the downside of inflationary pressure remains of course, at least when excess inflation is a risk during economic upturns. This is already modelled in the game, is it not? So there you have the downside. Inflation.

There are also questions relating to its effectiveness in stimulating the economy (happy to go further into this if you want me to) and critics argue that it drives inequality because it pushes up asset prices, making the owners of those assets, who tend to be wealthy, even wealthier. You could include that in the game as well if you want, by making QE increase inequality.

PS: Please include direct monetary financing! If you can print money for citizens (helicopter money), then you can print money for the treasury.

[cliffski]

All very interesting. I definitely think its true that QE pushes up asset prices, and basically makes investors wealthier, at the expense of the general taxpayer who will have to pay off the extra debt at some point in the far future. If we haven’t done so already, I should maybe link QE to generational wealth gap…

I understand the logic in the pushing up bond values thing, but here is where I get confused. Say as an investor, given the risk profile of govt bonds, I will buy them at £100, to get my 10% return. The government then buys up some of those bonds with QE, and the price goes above £100. Thats fine… but I don’t have to buy government debt of any sort, especially not a specific country. If government bonds become a bad deal, I can take the money I normally buy bonds with and invest in stock instead?
I thought that was exactly why asset prices rose during QE?

I probably know just enough about the topic to be dangerously confused (I did a degree in economics at LSE, but that was a long time ago!).

[Maphylius]

QE does not by itself create extra debt that would need to be paid off! Under QE the central bank creates additional reserves (“prints money”) to buy assets. It does not borrow this money! What is supposed to happen later on is that the central bank sells those assets to the private sector again and deletes the money again. In reality we haven’t really seen that happen too much since central banks have mostly just kept buying more and more assets, especially now due to the pandemic.

QE could increase government debt, only if the government decides to issue more debt as a result. Perhaps because they feel comfortable borrowing at the lower interest rate that results from QE. But QE itself does not increase debt that might have to be repaid with taxpayer money.

The government then buys up some of those bonds with QE, and the price goes above £100. Thats fine… but I don’t have to buy government debt of any sort, especially not a specific country. If government bonds become a bad deal, I can take the money I normally buy bonds with and invest in stock instead?
I thought that was exactly why asset prices rose during QE?

That’s right. If you are an investor like this you might happily sell your government bonds to the central bank at the higher price and then use that money for something else. That’s part of how QE is supposed to work. The hope is that you will use the money in a way that boosts demand, for example by buying a yacht or by investing in an expansion of some business operation. In reality, most of the money is used to buy stocks instead, which pushes up stock prices.

That’s why some people argue that QE is not as effective as for example Helicopter Money at boosting demand. Helicopter Money gives money to poorer people, which are much more likely to spend it (on something other than stock). Problem is that it is not reversible. With QE the central bank can remove excess money again by selling the assets back to the private sector. With helicopter money that’s not really possible.

I would be happy to have a chat about all this sometime if it helps improve the game.

[cliffski]

Aha, yes I mean increasing debt in the loosest terms (the game does not link QE to increasing debt).
I agree Helicopter money feels fairer, but my own opinion is that infrastructure spending is likely the best possible type of QE/monetary expansion. I do think that not enough people are really aware of how much QE is happening, and who benefits from it.

In the game, both QE and helicopter money have an effect on the credit rating of the country. This was reasoned because I assume that people are less likely to buy government debt if they think the government will effectively devalue (by printing money) to pay that debt off. I would suggest that QE/helicopter money are also both signs of an economy with issues, which should make people more wary of buying the debt of that country in general.

I always argue with people who claim that QE and money printing has no downsides, along the lines of ‘would YOU lend money to someone who is clearly planning on devaluing the currency they are paying you back in?’

[Maphylius]

In the game, both QE and helicopter money have an effect on the credit rating of the country. This was reasoned because I assume that people are less likely to buy government debt if they think the government will effectively devalue (by printing money) to pay that debt off

This is an interesting point. Ofcourse one would be less willing to accept a 2% return on bonds if you have 5% inflation eating away the value of what you will be paid back. But this should be an effect of inflation, not a direct effect of QE. The more direct effect of QE is definitely to lower interest rates paid by the government because it pushes up the price of bonds. I am not aware of any economist that doubts that this is what happens at least in the short term. The empirical evidence for this is quite clear and the Bank of England accepts this as well in their explanation I linked earlier. Of course in reality we have had low inflation despite QE for a very long time now. That’s why governments have generally had to pay very little interest.

So my suggestion for the game would be that inflation increases interest paid on debt, while QE lowers it in the short term.

my own opinion is that infrastructure spending is likely the best possible type of QE/monetary expansion.

Infrastructure spending is Government spending. So if you want to be able to do that in the game (which I would love aswell) you have to implement Direct Monetary Financing, i.e. printing money for the treasury. It would fit the whole “Zimbabwe” theme of the game as well. Zimbabwe did not use QE or helicopter money. They printed money for the treasury.

[cliffski]

very very interesting. I definitely need to take a look at my model for this.

[cliffski]

I’ve been looking at the code and data a bit, and was looking into making a link between inflation and credit ratings (which are basically interest rates, as expressed in the game), and wanted to add a link here.
Currently the code that works out what credit ratings should be checks if any ‘money-printing’ policies are in place (basically QE or helicopter money), and if they do, it then increases the debt risk and worsens the credit rating.

I was thinking that this is simply a hacky way of bypassing the inflation simulation.

Both QE and Helicopter money increase inflation already, so I’m thinking about scrapping this hard coded link between those policies and the credit rating, and instead just having a link between inflation and credit ratings. I’m thinking that:

1. This simplifies the code and makes it more flexible to equation changes.
2. There should be a general purpose link from inflation to credit ratings, regardless of the source of that inflation (could be strong unions pushing for wage rises as well).

Which sounds sensible to me, but then I still cannot get my head around this idea that QE reduces interest on government debt. I think we are debating two conflicting forces here:

• The effect of ‘cheap money’ pushing down interest rates in general (for everyone)
• The effect on the perceived stability and credit-worthiness of a government

I’m reckoning that the second outweighs the first, but I guess it could be that this depends on how much QE there is.

[Maphylius]

I think this is a reasonable first change to make. Though there might be a problem here with conflating “credit rating” and “interest rate on govt debt”. I am not sure what other effects the credit rating has in the game. I think it can cause an adverse event “debt crisis”? But you have to consider that high inflation does not mean a country is more likely to default. It simply means that creditors will expect a higher nominal return, because the money they get back will be worth less. It does not mean the government is at risk of default. In fact inflation tends to make it EASIER for governments to pay back debt, since it reduces the value of that debt.

Now to the effect of ‘cheap money’ under QE pushing down interest rates in general:

For the private economy, this should boost GDP, because it is eaiser to borrow money for investments. As far as I know this is already implemented, so no change needed here.

But for governments it also makes it cheaper to borrow. So QE should definitely reduce interest rates paid by the government.

I know you think there must be some kind of catch with QE. And there is the catch that it can cause inflation. But as long as inflation remains low, QE is relatively unproblematic in terms of stability of the economic system. I think you should trust the wisdom of central bankers around the world on this one, who have been relying on this heavily for a good reason.

[cliffski]

True. I see how it works, I get the impression that this is one of those things where the policy works fine…until it suddenly doesn’t, and a tipping point is reached. Unfortunately, true though such systems are, they make for a nightmare game design, because generally speaking in games you need to ensure a player has a lot of warning that something bad is likely to happen before you punish them

Currently the ONLY purpose in the game of credit ratings is to indicate changes to the interest rate on government debt, so in gameplay terms the two systems are basically the same.

[Maphylius]

Currently the ONLY purpose in the game of credit ratings is to indicate changes to the interest rate on government debt, so in gameplay terms the two systems are basically the same.

Great! No problem then.

generally speaking in games you need to ensure a player has a lot of warning that something bad is likely to happen before you punish them

The warning should be that inflation is getting dangerously high! In fact that is already something I look out for when playing the game and that is exactly what central banks always watch. For them it is always a game of balancing the growth of gdp with the need to control inflation around 2%. This is called the “dual mandate”. I would love to have a game where you have to do this.

[AvianOverlord]

@Maphylius

I know you think there must be some kind of catch with QE.

What if the catch is that it’s more efficient (in terms of GDP/inflation ratio) than helicopter money, but a lot less popular? Effective policy vs popular policy is uncommon in D4 but the gameplay fully supports it.

Edit: Generally, D4 is a game about governance, not politics. So polices tend to be balanced based on their effects, rather than on net popularity or capital cost.

[Maphylius]

I think you are right that the catch of QE is not that something is lost compared to where no measure is put in place, but the opportunity cost of implementing some other measure such as helicopter money.

No idea whether QE is more efficient in terms of GDP/inflation as you suggest.

[Tommy]

Another vote for Direct Monetary Financing! I feel that QE is a very capitalist money policy, while helicopter money is neutral, and DMF is the socialist version. I think adding it would give better balance.

[Tommy]

Here’s a good paper on how Canada successfully used DMF from the 1930s to the 1970s.

[PDF] Breaking the taboo: a history of monetary financing in Canada,...

[boreal_dragon]

i have an idea for the deflation issue, instead of having separate inflation/deflation combine them into one stat, if lower than 50% then you get inflation, if higher than 50 deflation, 50 is equilibrium.

[cliffski]

@Maphylius

In fact inflation tends to make it EASIER for governments to pay back debt, since it reduces the value of that debt.

Surely thats only true for people who are domestic holders of the debt? So someone who lends the government £10 and expects to get their £10 back doesn’t care as much as someone who lent the government £10worth of dollars, and will expect the current exchange rate value of $10 back?
(depending how such debts are structured and financed… I am no expert in this matter).

[Xenomorphman]

This would be generally correct, however there do exist some skewed wealth contracts IRL that create the scenarios Maphylius described, where the currency is outright stated in the deal, and when inflation hits it’s too late to modify the deal to be through another currency/ resource.

[cliffski]

It seems crazy to me that foreign investors lend money on such terms, but clearly they do!