Recent Posts

Friday, March 13, 2020

Yes, It's Not $1.5 Trillion Of Spending

There's been some complaining to-and-fro on Twitter about the $1.5 trillion repurchase agreement ("repo") facilities by the Fed. On the chance that any of my readers have run into this, just want to say a few words.

  1. It is a loan, not the same thing as new spending.
  2. Giving entities loans is commercially advantageous, and so it quite reasonably can be described as a "bailout." The trick is understanding that the magnitude of the bailout is not the same size as the dollar amount of the loan.
  3. This action underlines that the central government is the monopoly supplier of government money, and so this shows the firepower that alleged "bond vigilantes" actually face.
Repos have a legal structure that appears somewhat complicated, but their economic effect is best modeled as a loan against a bond used as collateral. These repos (unless I missed something) are against Treasury collateral. 

So, this is not the same thing as the Federal Reserve handing people money, or even buying stuff (although there was a portion of the package that are outright purchases). If you wanted to borrow $1000 in this fashion, you have to plop down about $1000 in Treasury bond collateral, and pay the loan back at the end of the repo term (possibly the next day).

This is very different from the central government handing you $1000 -- that is a form of income, that you are free to run off and spend. Realistically, that is what most government transfers will accomplish, and will have a much higher multiplier on economic activity. In other words, if the U.S. government sent out the equivalent to $1.5 trillion in cheques to everybody, that could easily have an inflationary impact. (Given the hole blown in economic activity, I do not have an idea how inflationary it would be.) Meanwhile, the repo market intervention will have almost no direct effect on activity; it just helps avoid the counter-factual of some form of market functioning disruption (that presumably would blow a hole in activity).

Nevertheless, this is not a truly neutral activity. If one had made financial arrangements with the friendly local loan shark, and you were somewhat short on the means of repayment, a government official popping in to give you a quick loan will save you your kneecaps. Your kneecaps have a definite value, but they may or may not match the dollar value of the loan.

More seriously, this intervention is helping out entities that were otherwise distressed. Yes, we need to prevent a collapse in financial intermediation, as we saw the side effects in the aftermath of the Lehman failure. However, we are forgetting part of Bagehot's prescription, as summarised by Paul Tucker:
to avert panic, central banks should lend early and freely (i.e., without limit), to solvent firms, against good collateral, and at ‘high rates'

Instead of backstopping liquidity by offering loans at penalty rates, the Fed is helping out financial firms that were on the wrong side of the market. One can argue that this is being done to avoid the dampening effect of higher interest rates on the real economy (as well as stabilising the Treasury market, which I view as the correct justification -- see below), but it also smells of "socialism for me, rugged individualism for thee." This does not sit very well with the free market absolutist ideology of most bankers, which every other sentient adult picked up on since the end of the Financial Crisis.

Finally, this episode underlines the arguments of neo-Chartalism (Modern Monetary Theory). Government money is an inherently valueless token, and the government is a monopoly supplier of it. The only issue is matching up money versus real resources. Those real resource constraints matter, not the arbitrary dollar amounts on the tokens.

Repos allow speculators to finance Treasury positions at the risk-free rate. So long as the Federal Reserve is consistent its dedication to preserving the Treasury repo rate near a policy rate, there is effectively unlimited firepower to buy bonds if yields are "too high." This unlimited firepower dwarfs the ability of "bond vigilantes" to "discipline" the Federal Government. This factor explains why I think the repo operations are good policy -- the Fed should maintain an orderly Treasury repo market, as it is critical for Treasury market functioning.

(c) Brian Romanchuk 2020


  1. I think the time sequence of events is something like this:

    Firm A has bonds but no cash; needs to pay firm B.

    Firm B does not NEED cash but wants cash. It probably already has bonds.

    CB steps in to buy bonds from Firm A; Firm A can pay Firm B.

    Firm B gets cash, does not NEED cash, wants bonds.

    Expected Future: Bond price should rise, thereby lowering interest rates.

    Possible outcome: Firm A recovers bonds from CB at agreed price. Firm A now has bonds to resell at slightly higher price. Firm B buys bonds from Firm A. Firm A has a small profit.

    Macroeconomic effect: Interest rates fall; cash in private sector slightly reduced due to small profit for CB.

    Brian: Would this sequence play out even if the two firms were private banks?

    1. I don't see your story as fitting what is happening. Financial firms of many types borrow against Treasuries to finance positions. They do this in the repo market. The Fed is stepping in to keep the market in repo financing orderly. Very limited implications for the price of bonds.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.