Let’s Talk About Shrinkage

There’s a lot of concern these days about shrinkage. No, not Seinfeld shrinkage. We’re talking about Federal Reserve Balance Sheet shrinkage and how the Fed might go about reducing the size of its balance sheet.

When the Fed started expanding their balance sheet in 2008 during the financial crisis there was a great deal of outcry and misinformation floating around. I wrote about 10 billion posts on this trying to clear up the misunderstandings. Let’s review the basics in case you missed it:

• During the financial crisis the Fed expanded its balance sheet to help stabilize the economy. They do this in the same manner that all banks expand their balance sheet – from thin air. This process involves the creation of reserve deposits that are then utilized to purchase assets from their partnering banks (many of whom had purchased these assets from other market participants).
• By expanding its balance sheet the Fed created super high quality short-term instruments and traded them in exchange for high quality long-term instruments like Mortgage Backed Securities and T-Bonds. As you might recall, some of these instruments were under price pressure in 2008-9 so this process helped to shore up the banking system significantly by improving the short-term liquidity and quality of the instruments held in the private sector. This, in my opinion, was undoubtedly beneficial as it helped stabilize a financial system in deep need of it.
• Importantly, what this process was not akin to was “money printing”. This is due to the fact that operations like QE do not actually expand the quantity of net financial assets in the private sector. In other words, the Fed created reserves and traded them to the private sector, but the Fed also removed a T-bond or MBS at the same time. So you could say that they printed a super short-term instrument into the private sector and unprinted a long-term instrument from the private sector. It can be helpful to think of the Fed’s balance sheet as being “off balance sheet” because it is not a balance sheet or income statement that directly impacts the real economy in the same way that a household or business’s every day spending does. In other words, the Fed doesn’t go shopping at Walmart so any assets accounted for on its balance sheet might as well be buried in a hole somewhere.¹
• As I predicted back in 2008 and 2009 QE did not cause high inflation, surging interest rates, high growth, and was not really all that impactful given all the fuss about it. Yes, I have argued that QE1 was probably very effective because it shored up balance sheets at a very unusual time, however, the future iterations of QE and the aggregate impact has been fairly small given how expansive the policy was. This makes sense since the Fed was essentially changing the composition of the private sector’s balance sheet rather than directly expanding it (as real bank loans do).  Swapping a savings accounts (T-Bonds) for a checking account (reserve deposits) doesn’t make anyone go out and spend more since it is little more than an asset swap of like assets.²

All of this resulted in the Fed owning a huge amount of long-term instruments that it earns interest income on (this is income the private sector doesn’t otherwise earn which could explain why QE has been marginally disinflationary).  Despite this, there is still tremendous confusion over what this all means. Some people think the Fed has to unwind this balance sheet, which, as I’ve shown before, is incorrect. The Fed can maintain control of interest rates by paying interest on reserves and just let the balance sheet mature over time which will naturally reduce the size of the balance sheet. Actively selling assets back into the private sector would achieve the same basic result since it would proactively extinguish the assets it created.

How does this process work? Well, it works in exactly the opposite way that expansion does. When the Fed creates reserves it expands its balance sheet from nothing and swaps the reserves with whatever it buys. If the Fed were to sell these assets back into the private sector the exact opposite would occur. The Fed would sell the T-Bond and obtain back its own liability thereby extinguishing it. Voila! Shrinkage.  See, that’s not such a big deal is it?  So don’t let all the talk about shrinkage get you down. After all, this type of shrinkage is very unlikely to be all that consequential.³

¹ – This is a bit of a generalization. In reality the Fed’s balance sheet does impact the real economy because it reduces the amount of interest income we earn. This further impacts fiscal policy since the Fed remits billions to the US Treasury every year.  

² – See my formal primer on QE if you want more detail here.  

³ – I would be remiss if I didn’t hedge this statement with the disclaimer that you just never know how market participants will respond to shrinkage. The behavioral risk is very real and the misunderstandings around QE are vast. But the real economic impacts are minimal as shrinkage, just like expansion, is a simple asset swap of like assets.