On the poor man's NGDP targeting regime

The main idea: issue, sell and buy unlimited quantities of consols whilst sterilising all resulting net stock changes for a bare-bones automatic NGDP targeting regime.  

0. Posts on this blog are ranked in decreasing order of likeability to myself. This entry was originally posted on 21.01.2025 and the current version may have been updated several times from its original form. See a similar system here.

1.1 Earlier I have written about a thought experiment used to model the fiat economy. I actually think that model is good enough that the real-world complications and ways in which reality differs from it are minor.

1.2 Minor enough that simplifying reality to fit the model would actually enable us to come up with a decent automatic NGDP targeting regime, absent complicated instruments like futures and whatnot.

1.3 So here it goes. You have a Monetary Authority tasked with three duties: husbanding foreign exchange reserves, bailing out failing banks and issuing and buying back consols, the last two of which are the only way it can interact with the local economy and create money.

1.4 The first duty is a familiar to any investment firm and of the second I’ve written about previously, so let’s focus on the third.

1.5 The Authority auctions off a number of consols that will pay out a constant stream of revenue every month forever. It is important that the value issued should be less than the foreign reserves stock for reasons that will become obvious. All receipts are reissued in the economy by buying out foreign currency.

1.6 Note how the auction sets the price of these instruments, and thus the nominal yield. 

1.7 The Authority now pledges to issue and buy back an infinite number of these instruments at the price that was discovered in the auction, forever (or thereabouts, see 1.14). Tasked with money creation, it can credibly pledge this. 

1.8 At this time, and assuming this promise is credible, the yield of these consols becomes the expected NGDP growth rate, immediately, with a huge shuffle of all intertemporal liabilities denominated in the local currency being set off to ensure this. 

1.91 Every fortnight, the Authority calculates the net number of consols traded in or out to sustain this pledge. It then counters this net flow by selling or buying an equivalent number of instruments, with all operations in this case effected in foreign exchange instead of the local currency. In a fortnight, we again calculate the net in- or outflow (ignoring our own earlier countering trade), and repeat, and so on. 

1.911 Alternatively, you could enforce a tripwire system where the net flow exceeding a specific number would trigger the intervention, or even combine both, for intervention triggered by either the tripwire or three months after the last. Anyway.

1.92 Originally, the sterilisation mechanism I came up with required the authority to sterilise the new inflows of money received or paid to the market and, whist this might look equivalent to countering the number of instruments traded it or out in the context of a set interest rate, it causes the monetary response to be backward and pro-cyclical.

1.93 For example, suppose the demand for money drops, and the public finds themselves with too much money on their hands. They will buy consols, resulting in a net inflow of money to the Authority. The former rule would have required the Authority to push this money back into the market by buying foreign exchange, thus working counter to the demand change.

1.94 Countering the change in outstanding instruments though would require the Authority to buy back the instruments that the market has acquired, which would now put foreign exchange in the hands of the market, having retrieved money balances earlier (note again that the consols being counter-traded are auctioned off in terms of foreign exchange, though the coupons themselves are of course still denominated in local currency).

1.941 This might ot might not trigger the market to buy some instruments again, which would cause another round, and so on until equilibrium is reached or demand changes.

1.10 Thus, the yield of the instruments becomes the expected as well as actual NGDP growth path. Every deviation from this rate translates into the market selling to or buying from the Authority, which trades are sterilised for an iterative and entirely automatic process that brings NGDP back to target. 

1.11 As long as foreign reserves comfortably cover the original issue of consols and then some, credibility will be fine. Any additional issue from then will mean inflows which will require sterilisation in the form of adding to our foreign reserves.

1.12 The only issue would be a major appreciation of our currency’s value in terms of those currencies held in reserve, such as to make the sterilisation of large and sudden enough stock changes impossible to honour. Whilst I’m not sure how such an exchange rate zoom would come about given our local yields are credibly fixed, let’s assume so. First of all, a nice problem to have.

1.13 Second of all, even such a cataclysm as above would simply mean that the second bit of the mechanism, the one where we sterilise stocks changes to make sure that expected NGDP growth translates into actual growth, breaks down. NGDP growth will continue to match our yield, but the base its growing from will be subject to this one random shock.  

1.14 Now, for a proper NGDP regime calque, set pre-determined check point at which the yield can be amended by pre-determined amounts. Say, the Monetary Authority (or, better still, Parliament) can change the yield (the price) by no more than ±1% every five to ten years. This is done to make sure than the rate evolves in reasonable lockstep with real GDP growth.

1.15 And there you have it, a nearly entirely automatic NGDP targeting regime and third world country can institute today.