Nouveau central banking, a proposal.

Every so often I like to put out a very 'blue skies thinking' blog on how to reform the tools and mandate of central banks. It's always based on the assumption that we want to give central banks some quasi-fiscal powers to avoid the zero lower bound problem and that we want to keep interest rates relatively constant if possible.

Now, you may quite reasonably say "hold on, I’m not sure we DO want those things!" Pretty reasonable retort there and I acknowledge your point. However, 1. This is blue skies thinking to examine what’s possible, 2. Yes governments could change how they operate fiscally to avoid the ZLB problem but let’s all admit national debt scares people and is a political problem and 3. is it not at least somewhat plausible that Keynes was correct in General Theory when he posited that messing about with interest rates was suboptimal because it might block useful investments during the high inflation times that would expand capacity (and thus lower inflation) and also that predictability is really important to business and so stable interest rate expectations coupled with stable inflation expectations might reap rewards?

I hope you'll at least agree it's an interesting hypothetical so, let's play with it.

Let's start by acknowledging something. I don't intend to take the MMT position here and I do support the idea that the base underlying mechanism should remain as it is now. I.e. if these additional tools are in place and inflation is still rising then I would continue to advocate for the current normal policy response of raising overnight interbank interest rates. I would, however, for the purposes of this discussion, shift the target inflation rate to a range of 2% to 4%. That’s mostly so we can feel confident that the economy really is firing on all cylinders. It's also to try and ensure a negative real rate for safe long run rates of return. This is a core part of this whole endeavour and is intended to address the problem of money generating more money at zero risk. It seems very hard to justify why, in an ideally perfect and beautifully competitive market, we should be paying rich people just for being rich without them taking on risk. That seems distortionary and yet we accept this state of affairs most of the time. That seems wrong both from a market competition point of view and a moral point of view.

That brings us to setting our aim for the long run risk free rate. After all if it's not going to jump about very much then we can now set where we want it to be with the aim of adjusting other policy levers to push r* to our target long run rate. My proposal here is 2%. This is lower than the rate I've previously suggested, 3%. That’s because 2% makes it easer to hit the negative real rate of return on risk free assets. The counterargument is that assuming the risk stays the same then the smaller the risk free rate of return is the greater the apparent gap in borrowing rate between big low risk businesses and small, dynamic but risky businesses. That is because if the risk stays constant and the risk free rate shrinks then the risk part of the interest those businesses have to pay is an increasingly large proportion of the total interest paid. That has the worrying possibility of reducing market competition. However, 1. The difference between 2% and 3% is rather small, 2. You can't go much higher if you want to have a negative risk free rate of return and also have relatively low rates of inflation and 3. I think there are other ways you can advantage small businesses in the tax code to compensate (and to be clear I think this is incredibly important and we should do it.)

OK so now we have our long run risk free rate we are going to add our first new central bank tool. We're going to have our central bank issue annuities (or you can consider them consuls if you prefer). We have now jumped into the central bank doing fiscal policy looking things, however, it's not as scary as it first seems. After all banks issue debt all the time and they certainly take deposits. It’s perfectly normal for banks to offer vehicles for people to save money in, this is no different. So the central bank will issue, on demand, in helpfully small denominations of £100, 2% annuities/consuls. These will be available through "inner banks" i.e. those with accounts at the central bank. Further these inner banks will be required to follow standards set by the central bank to hold certain amounts of these annuities themselves, presumably via some sort of reserve requirements based on deposits they have, such that the market value of these annuities can be pushed towards their nominal value. This is, of course, a bit of a pain for those inner banks but that's the price they pay in return for the large reward that comes with being inner banks with access to central bank reserve accounts.

Keen observers might be wondering, at this point, what the point of all that is. This system won't control inflation (unless reserve requirements get to 100% or more in which case lending/money creation becomes far more difficult), true. However, it does set our long run risk free rate in a tradable instrument. That’s important because we want economic actors to be able to switch in and out of that instrument such that it becomes the benchmark for that long run risk free return. That's deeply relevant for our next tool which is considerably more like real fiscal policy. That is a land value tax paid directly to the central bank.

Economists for centuries seem to have agreed that LVT is pretty unequivocally good as taxes go. A reasonable question might be what is the optimal rate for it? My proposition here is that if land values sometimes fall then mortgages can get into negative equity and that can cause economic chaos, equally if land values rise rapidly that can distort markets by transferring wealth to land holders, even though only some of them can access that wealth because only some of them are holding additional land beyond their primary residence. Therefore just as the monetary policy committee currently moves interest rates around to target general inflation I would propose that the LVT rate is moved around to control land value inflation. (So that this is politically possible I suggest the first £200,000 of land value of one property can be exempt from tax per person. Yes that's suboptimal economically but otherwise it would be politically impossible to bring in LVT.) The target for land value inflation should be for median land values nationally to be between 1% and 2%. This gives, in my view, the greatest likelihood that few properties will go into negative equity on their mortgages as if the mean was used instead this would be highly distorted by high value land.

The connection between these first two policies is important. If the genuine risk free rate of return is 2% provided by the central bank with an instrument that trends towards par due to those reserve requirements then the value of land is intrinsically linked to that amount. With LVT being raised and lowered such that the median value of land rises somewhere in the 1% to 2% range any individual plot of land, and whatever buildings are attached to that deed, will have some expected return via those land value rises and rents and some risk attached. The market expectation for those returns minus the market expectation for that risk needs to net out to that 2%. This is because if this net value is lower than 2% it makes sense to sell the land and buy the central bank annuities. If it is higher then it makes sense to sell annuities and buy the land.

So what should happen if inflation rises? Well here we must remember what we mean by inflation in central banking, we are specifically referring to the idea that there is an oversupply of the currency/loans and that is swamping available supply. In that scenario the goods we would most expect to rise in price are inelastic goods. There is no good more inelastic than land and, helpfully, it is also a vital good. So in the vanilla inflationary environment central banks normally operate in (important caveat there) these two policies together should dampen inflationary swings. During the boom times pressure should push land values upwards and if the long run expected risk free rate of return hasn't moved, because of the annuities, this should very clearly be seen in land values. Land, just like the annuities, is an effectively infinitely lived asset so long as the institutions that govern the country continue to exist (the central bank and the current land ownership laws). This push forces the central bank to raise the LVT. This takes money out of individuals and businesses pockets and, if the central bank does not replace them, reserves out of the banking system. (We'll come back to that in a moment.) Conversely in a normal recession we would expect the reverse, a stagnation or fall in land values, causing a drop in LVT at the same time as a flood into the annuities. The drop in LVT would then decrease expenses for individuals and businesses and help give an automatic stimulus to the economy.

So, to address the points that come out of that: 1. If individuals don't pay LVT on the first £200,000 of their primary residence surely they won't see any change? Plus renters won't either? Yes, absolutely true. However those same individuals probably aren't in a position to take advantage of a change in interest rates under the current system either. Monetary policy can still be effective even if it isn't perfectly targeted, it's just a little slower. (Plus there's going to be some extra tools later on that change the game here). 2. OK so what about reserves then, I thought we weren't changing the overnight rate? Good point. My proposal would be this: we probably don't want the overnight rate swinging too much because we don't want liquidity issues, however, there's no reason the spot rate has to always match the long run rate. I would therefore have the central bank operate a ceiling on overnight rates of, say, 4%. Otherwise let reserves bounce around. This will allow the policies above to drive the short run rate towards supporting the policy objective. I.e. it should rise during booms and fall during busts. Of course should it be necessary to raise this cap due to inflation failing to be controlled yes, obviously, that is preferable to hyperinflation. If the central bank needs to set a floor on interbank lending rates it can always pay interest on reserves. The point is that this will usually not be necessary. 3. Ok but other than the 4% cap how are we injecting money in? Glad you asked.

The third and final fiscal tool is the companion to that land value tax. The LVT is the taxation, the annuities are the "borrowing" so what is the spending? Well, naturally, a UBI. I'm going to suggest a reasonably low starting point of £100/week (that's £5,200/year). That’s not enough to meet the UK poverty line of £150/week but we'll come back to that. The idea here is that in the first year or so the LVT and UBI should roughly match so as to not alter the balance of money in the economy too much, although obviously this will definitely result in redistribution! It may be worth phasing both in if necessary. After it is established though the UBI would rise at a set rate of 4% per year. This should only be altered very rarely and via legislation on advice of the central bank due to clear, endemic, long run factors. The explicit aim is that a society should aim to have constantly rising standards of living for all so 4% is at the top of the target inflation range. When added to the LVT taking money (including reserves) out this UBI injects cash (including reserves) in.

I should add that the existing welfare state provided by the government through the treasury is unaffected at this point and so most people should now be above the poverty line. Of course governments may then begin to withdraw parts of the existing welfare state they no longer consider necessary.

The idea for this first half of the proposal is to set a baseline set of economic conditions which should, generally speaking, smooth out the boom and bust cycle without having swings in the long run expected rate of return. Generally speaking in boom times with higher profits returns will be somewhat muted by higher land costs and mildly higher borrowing rates. Conversely in the bust times incomes should be helped by lower land costs and cheaper immediate borrowing.

However, to really make sure of the borrowing policy lever we need to step out from stealing a march on fiscal policy and return to the more natural space for central banks- lending, borrowing and savings.

The policies above, because they include the current policy set as the base, can theoretically cover any scenario current policy can, however we want to try and make sure that central banks don't need to use the current tools so we want to give them a few more tricks. That's where the fact the central bank is a government entity becomes useful.

Traditionally lending, at least outside of the Islamic tradition, involves some principal amount lent for some period of time and then interest paid as a coupon on top (in a home mortgage of course these are paid together but the point is the same). The UK, however, regularly uses an entirely different tool that we can reimagine as a useful way to create a whole new toolkit for central banking, our student loan system.

First, the proposal as it relates to individuals (presumably UK citizens): you as an individual would have the option in any given year to take out one student loan(ish) style loan (I.e. a loan paid back via a tax/a set percentage of your income) service from the central bank via an inner bank. In addition inner banks would be given access to the repayment system and so could offer competing options if they felt that was good for them. Once you had taken out a loan you would get an initial kinder repayment tax of 5% but if you did then take that option again and have two or more loans running simultaneously then you would pay a less generous 10% tax. This is essentially just to ensure people quench any feelings that the system is too unfair either by being too harsh right away or too generous to repeat use. This does obviously mean that from the third loan onwards (until you repay the amount) you only ever have to pay the 10% tax and the additional money isn't adding to your yearly payments and effective interest rate but it instead adds to the duration of time you'll have to spend paying back the loan.

The amount paid back would deviate from the current student loan system however by you paying back a flat 1.25 times the amount you borrowed. This means that how good a deal this facility is depends very specifically on how quickly the loan would be paid back via that percentage of your income. If it takes longer then the effective interest rate is lower, if it's quicker than the effective interest rate is higher. Therefore if you have a larger income paying the same percentage on the same size of loan it will be paid back quicker and therefore at a worse effective interest rate. It will therefore tend towards being a bad deal for people on larger incomes and more likely to be a good deal for those on lower ones.

The cap on the amount that can be borrowed is the policy lever for the central bank, why would this work as a policy lever?

Well first consider who would find these loans useful? They are unconditional, anyone can take them out but whether they're a good deal depends enormously on the amount borrowed and what your income is. For example, if you borrow £4,000 you will have to pay back £5,000. If you pay that back on a £100,000 a year income you'll pay it all back in a single year. That's a 25% interest rate! It seems rather unlikely that the best interest rate available to someone earning £100,000/year borrowing just £4,000 is 25%, it's an obviously terrible deal. Conversely if you earn just £10,000/year you would pay £1000/year. That means you pay back in 5 years. That's an effective interest rate of 5% a year. Not great but considerably more tempting, especially if you really need it and you know that you only have to repay based on your income, not a set amount.

To consider a less extreme example imagine a £9,600 loan. This gets times by 1.25 to give a repayment amount of £12,000 exactly. A £30,000 income (around the UK average) that doesn't get any rise over the repayment time repays the loan in 4 years for an effective interest rate of 6.125%. A £20,000 income does it in 6 years, ~4.17%. Finally a £10,000 income takes 12 years, a very generous 2.08%.

So the system would work like this: everyone would get a minimum allowed borrowing amount of £3,000 and that would rise at 4% a year automatically, just like the UBI. Additionally you would be able to borrow a percentage of the amount you had paid in National Insurance during your lifetime plus your current holdings of those central bank annuities. This benefits pensioners the most as they would have the most accumulated lifetime NI contributions (although inflation means the contributions further back would, of course, be worth less today) I would therefore suggest that to make this disparity less extreme any contributions in the last five years should count for double. During times when inflation was above target the central bank would reduce the percentage of lifetime NI/holdings of annuities you could convert into borrowing. When inflation was too low it would increase this amount. What this does is it changes who finds it makes sense to borrow and increases/decreases the amount those people can borrow. It does this in a way that’s relatively progressive. NI is a regressive tax and so the wealthiest will not see their borrowing allowance go up as much as a percentage of their normal spending power as others, nor will it be as beneficial to them as they will pay a much higher effective interest rate. It makes annuities more useful because they are connected to this facility and it has the useful benefit of giving a gift to pensioners that doesn't really cost the young. Getting policies enacted often involves giving something to older voters who are more likely to vote. This policy helpfully doesn't directly cost the taxpayer anything because it all comes from the central bank to loosen monetary policy. It helps poorer pensioners out considerably because they're less likely to ever have to pay it back because the amount they borrow will be higher than the minimum, because of any NI over their lifetimes, while their future incomes will be lower. Higher income pensioners may well decide it isn't worth it because the tax incurred would fall on their private pension or investments and be too expensive to be worth taking the loan. Equally it may be that some middle income pensioners have relatively high amounts of lifetime NI contributions but relatively low future incomes and so will take out the loan and invest it rather than spend it. That’s fine too, that's simply the equivalent of looser monetary policy in low inflation times, exactly what we want. For younger people the deal isn't as good but it's by no means bad. Most people like knowing their parents and grandparents are doing well, many may even get financial help from them in these circumstances, but the young will benefit directly too. Those at the lowest end, most struggling during a recession, will have access to immediate cash with repayment only if and when they get income. It's very much an automatic fiscal stimulus.

How it would work for companies: for incorporated entities there would be no base amount. Instead company allowances would be based on how many annuities they held. The tax element would need to be far lower to make this viable for business so I suggest 1% of revenues if they take out only one loan and 2% thereafter. (I would consider also enacting a tax free allowance that each citizen could choose to attach to a company that would exempt the company from tax for that amount. This would then allow us to institute a rule where annuities of a value up to that tax allowance would count for double in terms of lending allowance. Additional annuities would count as normal. This is specifically to give an advantage to small business. It works because big businesses are not going to be able to get citizens to hand over their business tax allowances to them easily enough to get enough allowances to cover a significant part of their income. Small business on the other hand are often family run or embedded in the community and so can turn to employees/owners/the local community for support. This may seem a little convoluted but I really do believe it's important to counteract those worries earlier around the need to help small business to ensure a vibrant and competitive market! Obviously it isn't a requirement for the loan system to work but it will make it significantly easier for small businesses to access and use it effectively.) The percentage of annuity value that could be borrowed via the scheme would then be the same as for individuals and the central bank would alter this percentage as a policy lever. This does leave the question of why would businesses be holding these annuities that have a negative real yield?

Well firstly some businesses will hold them just because they reduce overall risk so where the company would otherwise be holding cash it may well often be better to hold annuities. Secondly, obviously, it can just be to access this facility if and when times do get tough! However, there is an additional reason. You can't really make individuals do forced saving, however, any incorporated entity exists specifically as part of a contract with a state, that's what incorporation is. I think it's reasonable, therefore, just as we tax businesses, to swop the shifting interest rates environment for a shifting required saving environment. Unlike with individuals we can’t have any advantage for older businesses so I would suggest that the central bank should set a requirement for any business to hold X% of its revenue for the previous year in annuities this year. That way all businesses should have some annuities. (Again I'd implement the only needing to do this 50% of the value of the tax free allowance individuals have applied to that business and then revenue after that is affected as normal. That, again, is to help small businesses and avoid creating an additional liability they can't afford.) To know what to set this level at I'd suggest the central bank look at stock market indices. If stocks are shooting up far faster than the risk free rate then the requirement should be raised, conversely if they’re falling it should be lowered. This hopefully trims the risk during the down times in exchange for shifting some company spending into safe assets during the good times.

Once you add these new loans for individuals and companies, plus the forced savings for companies, you should end up with a significantly more stable economic system without losing any dynamism or competitiveness. It's unavoidable that this looks more interventionist than the current system and, on the face of it, it is, but its overall effect is not actually particularly more interfering than before. Yes it does involve things like forced saving, absolutely true, but it hopefully removes the sharp cliff edge of being a company that could borrow for years with no problem but then suddenly finding yourself just as profitable as before but suddenly bankrupt because interest rates have shot up. In order to assess this proposal fairly I would submit that we need to acknowledge just how devastating and unnecessary the "credit crunch" of the post financial crisis era was. Hopefully between all these proposals, fiscal and monetary, that just wouldn't happen. With the drop in financial markets we saw at that time, following on from the previous boom in them, businesses would find themselves in a situation where their required holdings of annuities which had been going up during that boom had suddenly gone down. Meanwhile the amount they were allowed to borrow against those annuities (in exchange for a percentage of future income, not a fixed payment!) had gone up. They would therefore be able to either sell them, with the prices backed up by the central bank making sure inner banks bought the number necessary to keep the value of the annuities at par, or else borrow on terms that were considerably less risky than taking on fixed debt repayments during uncertain times. All while their customers were being helped out with income stabilisation policies. That would have significantly dampened the damage. Along with more stable interest rates this should mean the recession would have played out in a more orderly, less damaging manner without affecting competition and market forces.

On the other hand, if we were to consider the hyperinflation of the 1970s: the central bank could lower the percentage of NI/annuities value that could be borrowed against, to zero if necessary, as well as raising the forced savings for business and, presumably, seeing a rise in LVT all before having to use blunt interest rates (although clearly they still could). The LVT and the drops in the borrowing allowance against annuities/NI would have hit the better off harder than those with the least and would have minimised how much the damage caused by the episode had to fall on the backs of the poorest. Considering the extremity of the situation presumably some pain would still have to be experienced by all but having it more evenly distributed would, hopefully, have lessened the political consequences. Of course the real hope is that even with just today's tools a 1970s style episode could never happen but with all these extra tools layered on top it should be even more unlikely or even impossible.

All this while ensuring an unconditional and rising income for the poorest with flexibility made available when times get hard. Imagine those loans kicking in automatically during Covid, for example. That surely would have provided automatic relief far quicker and far more evenly than the actual government response. In addition, of course, putting all this on the central bank balance sheet makes it feel far safer to observers and removes much of the political tension that comes with national debt, deficits and fiscal policy.

One final though is to consider how the Eurozone crisis would have been different if they had had this policy after the financial crisis. Each EU citizen would have received the basic income and access to the loans (clearly NI is not something that exists in the same form in the Eurozone countries so we would have to image some equivalent!). LVT would have been spread based on land values and so those poorer countries that were so disadvantaged by having the same currency would suddenly find the opposite is true. They would be receiving the same support per citizen but their costs paid in, via land values, would be considerably lower. Those rich countries that so benefitted from having the same currency as their less well off neighbours would automatically be paying their fair share in, via the LVT, while only receiving the same € amount of support per citizen. The inequality and unfairness caused by the monetary union would suddenly be no issue at all.

So, in conclusion, this is obviously not a remotely practical proposal politically. This is not going to happen. However, I hope I gives some interesting food for thought about what is possible economically and that it makes some readers wonder if we couldn't steal just a couple of ideas from it here and there to make the world a little better, LVT with a primary residence tax free allowance, UBI and the loan system are all, I think, good and helpful ideas in and of themselves. Thanks for reading.

(P.s. if you’re wondering what this leaves as the absolute income floor: if you have no NI contributions at all over your life and if you’re forced to take out the loan in both year 1 and year 2 to cope then you’re paying 10% of your income in year 2. The UBI year 1 was £5,200. Increased by 4% to year two gives £5,408. So that minus 10% is £4,867.20 The minimum borrowing amount in year 1 was £3,000 so the plus 4% in year two is £3,120. Add those together and you get £7,867.20 a year or £151.29 a week. Just over the current poverty line before any other benefits get added. If current benefits are added then if someone gets £90/week universal credit withdrawn at the current 63%, and the UBI counts for that withdrawn but the loan does not, then this amounts to another £24.48 a week after that withdrawl. So a new total of £177.77 a week or £9140.16/year. Of course prices will also have gone up so the poverty line will have risen. However, even if the government starts to pull benefits away, the minimum amount you can be on should push above the poverty line and keep rising, assuming inflation on the things the poorest are actually buying can be kept below 4%.)