The monetary transmission mechanism in a 'perfect world'
I fundamentally think that what really sets Market Monetarism aside from other macroeconomic schools it how we see the monetary transmission mechanism. I this blog post I will try to describe how I think the monetary transmission mechanism would look like in a ‘perfect world’ and how in such a perfect world the central bank basically would do nothing at all and changes in monetary conditions would be nearly 100% determined by market forces. Futures based NGDP level targeting – the perfect world No monetary regime is perfect, but I think the regime that get closest to perfection (leaving out Free Banking) is a regime where the central bank targets the nominal GDP level and implement this target with the use of an NGDP-linked bond. How would this work? Well imagine that the government – lets say the US government – issues bonds linked to the NGDP level. So if the market expectation for the future NGDP level increases the price of the bond increased (and yields drop) and similarly if the NGDP expectation drops the bond price will decline. Now imagine that the central bank announces that it will always buy or sell these bonds to ensure that the expected NGDP level is equal to the targeted NGDP level. Then lets now imagine that the price of the bond rise is reflecting expectations for a higher NGDP level. If the expected NGDP level increases above the targeted NGDP level then the central bank will “automatically” go out and sell NGDP-linked bonds until the price is pushed down so the expected NGDP level is equal to the targeted level. This means that the central bank will automatically reduce the money base by a similar amount as the amount of bond selling. The drop in the money base obviously in itself will contribute to pushing back the NGDP level to the targeted level. It don’t take a genius to see that the mechanism here is very similar to a fixed exchange rate policy, but the outcome of the policy is just much better than what you would get under a fixed exchange rate policy. And similarly to under a fixed exchange rate regime the money base is endogenous in the sense that it is changed automatically to hit the NGDP target. There is no discretion at all. Changes in money demand will do most of the job It is not only the supply of money, which will be endogenous in a perfect world – so will the demand for money be. In fact it is very likely that most of the adjustments in this world will happen through changes in money demand rather than through changes in the money base. The reason for this is that if the NGDP targeting policy is credible then investors and consumers will adjust the demand for money to ‘pre-empt’ future changes in monetary policy. Hence, let imagine a situation where NGDP growth for some reason start to slow down. This initially pushes market expectations for future NGDP below the targeted level. However, this will only be short-lived as forward-looking investors will realise that the central bank will start buying NGDP-linked bonds and hence increase the money base. As investors realise this they will expect the value of money to go down and as forward-looking investors they will re-allocate their portfolios – buying assets that go up in value when NGDP increases and selling assets that go down in value when this happens. Assets that go up in value when NGDP expectations increase includes shares, real estate and of course NGDP-linked bond and also the national currency, while regular bonds will drop in value when NGDP expectations increase. This is key to the monetary transmission mechanism in the ‘perfect world’ – it is all about consumers and investors anticipating the central bank’s future actions and the impact this is having on portfolio reallocation. Similarly there is also an impact on macroeconomic variables due to this portfolio reallocation. Hence, if NGDP drops below the targeted level then rational consumers and investors will realise that the central bank will ease monetary policy to bring NGDP back on track. That would mean that the value of cash should be expected to decline relative to other assets. As a consequence consumers and investors will reduce their cash holdings – and instead increase consumption and investment. Similarly as monetary easing is expected this will tend to weaken the national currency, which will boost exports. Hence, the “NGDP anchor” will have a stabilizing impact on the macro economy. Therefore, if the central bank’s NGDP targeting regime is credible it will effectively be the market mechanism that automatically through a portfolio reallocation mechanism will ensure that NGDP continuously tend to return the targeted NGDP level. We can see in the ‘perfect world’ the money base would likely not change much and probably be closed the ideal of a ‘frozen money base’ and the continuously adjustment in monetary conditions would happens by changes in the money demand and hence in money-velocity. It should also be noted that the way I describe the transmission mechanism above interest rates play no particularly important role and the only thing we can say is that interest rates and bond yields will tend to move up and down with NGDP expectations. However, the interest rate is not the policy instrument and interest rate is just one of many prices that adjust to changes in NGDP expectations. The Great Moderation was close to the ‘perfect world’ The discussion above might seem somewhat like science fiction, but in fact I believe the way I describe the transmission mechanism above is very similarly to how the transmission mechanism actually was working during the Great Moderation from the mid-1980s to 2007/8 particularly in the US. Effectively the Fed during this period targeted 5-5½% NGDP growth and that “target” was highly credible – even though it was never precisely defined. Furthermore, the NGDP “target” was not implemented by utilizing NGDP-linked bonds and officially the fed’s used the fed funds target rate to implement monetary policy. However, the reality was that it was the market that determined what level of interest rates that was necessary to hit the “target”. Hence, only very rarely did the fed surprised the market expectation for changes in the fed fund target rate during that period. Furthermore, it was basically a portfolio reallocation mechanism that ensured NGDP stability – not changes in the fed funds target rate. So when NGDP was above ‘target’ investors would expect monetary tightening – that would cause market interest rates rise, stock prices to drop and the dollar to strengthen as future monetary tightening was priced in. In this process the demand for money would also increase and hence the velocity of money would decline. So the real achievement of monetary policy in the US during the Great Moderation was effectively to create a credible NGDP targeting regime where monetary policy basically was market determined. The problem of course was, however, that this was never acknowledged and equally problematic was the reliance on the fed funds target as the key monetary policy instrument. This of course turned out to be catastrophic defects in the system in 2008. In 2008 it was very clear that NGDP expectations were declining – stock prices was declining, bond yields dropped, the dollar strengthened and money velocity declined. Had there been a futures based NGDP targeting regime in place this would likely have lead to the price of NGDP linked-bonds to drop already in 2006 as US property prices peaked. As the fed would have pledged to keep NGDP expectations on track this would have led to an automatic increase in the money base as the fed would have been buying NGDP-linked bonds. That would have sent a clear signal to consumers and investors that the fed would not let the NGDP level drop below target for long. As a consequence we would not have seen the massive increase in money demand we saw and even if it that had happened the supply of money would have been completely elastic and the supply of dollars would have risen one-to-one with the increase in money demand. There would hence have been no monetary contraction at all. Instead the system ‘broke down’ as the fed funds target rate effectively hit the Zero Lower Bound (ZLB) and the fed effectively became unable to ease monetary policy with its preferred monetary policy instrument – the fed funds target rate. Obviously in the ‘perfect world’ there is no ZLB problem. Monetary policy can always – and will always – be eased if NGDP expectations drop below the targeted NGDP level. Fiscal consolidation in the ‘perfect world’ In the ‘perfect world’ the fiscal multiplier will always be zero. To understand this try to imagine the following situation. The US government announces that government spending will be cut by 10% of GDP next year. It is pretty obvious that the initial impact of this would for aggregate demand to drop. Hence, the expectation for next year’s NGDP level would drop. However, if NGDP expectations drop below the targeted level the fed would automatically expand the money base to ‘offset’ the shock to NGDP expectations. The fed would likely have to do very little ‘offsetting’ as the market would probably do most of the work. Hence, as the fiscal tightening is announced this would be an implicit signal to the market that the fed would ease monetary policy. The expectation of monetary easing obviously would lead to a weakening of the dollar and push up stock prices and property prices. As a consequence most of the ‘offsetting’ of the fiscal tightening would be market determined. We should therefore, expect money demand to drop and velocity increase in response to an announcement of fiscal tightening. As an aside it should be noticed at this is the opposite of what would be the case in a paleo-keynesian world. Here a tightening of monetary policy would lead to a drop in money-velocity. I plan to return to this issue in a future post. The important point here is that in the ‘perfect world’ there is no room or reason for using fiscal policy for cyclical purposes. As a consequence the there are no argument as consolidating fiscal policy is long-term considerations necessitate this. Market Monetarism is not about ‘stimulus’ and QE, but above rules I think my conclusion above clearly demonstrates what is the ‘core’ of Market Monetarist thinking. So while Market Monetarism often wrongly is equated with ‘monetary stimulus’ and advocacy of ‘quantitative easing’ the fact is that this really has nothing to do with Market Monetarism. Instead what we are arguing is that monetary policy should be ‘market determined’ by the use of targeting the price of NGDP-linked bonds. In such a world there would be no ‘stimulus’ in the sense that there would be no need for discretionary changes in monetary policy. Monetary conditions would change completely automatically to always ensure NGDP stability. As a consequence monetary conditions would likely mostly change through changes in money demand rather than through changes in the money base. Therefore we can hardly talk about ‘QE’ in such a regime. So why have Market Monetarists then seemly supported quantitative easing in for example the US. Well, the point is first and foremost that the fed’s monetary policy regime over the past five years have not been entirely credible – we are getting closer, but we are very far away from the ‘perfect world’. Hence, the fed needs to undertake quantitative easing to demonstrate first of all that it can indeed ease monetary policy even with interest rates basically at zero. Secondly since monetary policy is not credible (countercyclical) changes in money demand will not happen automatically so the fed will instead have to change the money base. Obviously these measures would not be necessary if the US Treasury issue NGDP-linked bonds and the fed at the same time announced an NGDP level target and utilized the NGDP-linked bonds to hit this target. If such a system were credibly announced then it would be very hard to argue for ‘monetary stimulus’ and quantitative easing in the discretionary sense. It might be that the discussion above is pure fantasy and it is pretty clear that we are very, very far away from such a monetary policy regime anywhere, but I nonetheless think that the discussion illustrates how important it is for monetary policy to be rule based rather than to be conducted in a discretionary fashion. Both the Bank of Japan and the Federal Reserve have within the last six months moved (a little) closer to the ‘perfect world’ in the sense that their policies have become a lot more rule based than used to be the case and there is no doubt that the policies are ‘working’. Especially in the case of Japan it seems clear that ‘automatic’ adjustments in money demand is going to play a very key role in achieve BoJ’s 2% inflation target. Hence, it is likely that it will not be the expansion of the money base that will do it for BoJ, but rather the likely sharp increase in money-velocity that will ensure that BoJ’s hits its target. Finally, I would argue that my discussion above also demonstrates why a proper NGDP level targeting regime is a true free market alternative as the system relies heavy on market forces for the implementation of monetary policy and is strictly rule base.