Yes, QE Creates Wealth Effects
Listen to the Audio Mises Wire version of this article.
QE (quantitative easing) is the primary means that the Fed uses to cause asset price inflation. "The Ben Bernank," Fed governor at the time, later chair, in an October 2003 speech endorsed the use of monetary policy to create wealth effects.
easier monetary policy not only raises stock prices; as we have seen, it also lowers risk premiums, presumably reflecting both a reduction in economic and financial volatility and an increase in the capacity of financial investors to bear risk. Thus, our results suggest that easier monetary policy not only allows consumers to enjoy a capital gain in their stock portfolios today, but it also reduces the effective amount of economic and financial risk they must face. This reduction in risk may cause consumers to trim their precautionary saving, that is, to reduce the amount of income that they put aside to protect themselves against unforeseen contingencies. Reduced precautionary saving in turn implies more spending by households.
In central banking doctrine, wealth effects are one of the channels by which Fed policy affects macroeconomic activity. The wealth effect occurs when monetary policy causes asset prices to rise relative to consumption goods prices. Owners of assets perceive that as an increase in their wealth. At the margin they will save less, perhaps even dissave and spend more on consumption goods. In the Keynesian framework, consumer spending drives an increase in production. Therefore, according to Keynesians, wealth effects have a beneficial impact.
Michael Sproul, commenting on the ThinkMarkets blog, disputes the idea that QE creates wealth effects. This article will show that both Bernanke and Sproul are wrong, in different ways, about QE.
Sproul’s argument is as follows:
The person who receives $100 from the Fed simultaneously hands over a $100 bond to the Fed, leaving his net worth unaffected. At the same time, the Fed’s newly acquired $100 bond provides backing for the $100 of newly-issued cash, so the value of the dollar is unaffected, and there is no possibility of wealth effects.
To pinpoint Sproul’s error, we will undertake a step-by-step exposition of QE. Let’s start from an a plain state of rest, meaning that everyone prefers what they have over any possible exchanges:
What value does Henry place on his portfolio? We can infer that he holds the bond because he values it more than the best offer he could obtain for it. Suppose that Lloyd has made an open offer of $93 for the bond. Then Henry values the bond at no less than $94, or he would sell it. We are going to assume that Henry’s reservation price for the bond is $95.
The Fed is also in equilibrium, meaning that it values its cash balances more than the asking price of any assets that it could purchase. The Fed would have to offer $96 or more to buy Henry's bond. By assumption, the Fed values $96 more than the bond.
In step 2, the Fed undertakes an ambitious QE program intended to cause asset price inflation, wealth effects, and then stimulation of consumer spending. A normal participant would only start spending and could stop holding cash due to a change in preferences. The Fed can do that, or it can create new money. Let’s look at the cases where the Fed creates out of nothing $100 in cash to finance its QE program.
Now the balance sheets look like this:
The Fed has had an increase in cash balances. Unlike all other monetary transactions, in which a seller must offer a good or service in exchange for money, and the buyer decreases his cash balance by the same amount, in this case no one has decreased their cash balance anywhere in the community. The total amount of money has unilaterally increased.
We are now no longer in a plain state of rest. Because the Fed has more cash, it values the additional $100 less than the first $100. The Fed now has an excess cash balance. It no longer values the marginal $94–$100 less than Henry’s bond. The Fed and Henry quickly agree on a price of $96 for the bond. The price must be more than the $93 offer from Lloyd, otherwise Henry would have already sold it, and it must exceed Henry’s $95 reservation price.
Central banking plumbing may sequence events in the opposite order—the Fed might buy the bond, write a check, cash the check, and then create the deposit. The sequence shown here is clearer for pedagogical purposes. Both timelines rely on the Fed’s ability to buy an asset with money that it creates. The Fed can buy assets with money that it either does not have or does not wish to spend out of existing cash balances, knowing that it will create the money with which to pay for the asset when the check is returned.
This is also unlike gold mining in a gold-based monetary system, where the miner increases their cash balance by extracting gold from the ground. The cost of extracting a single ounce—consisting of cement, energy, trucks, shovels, crushers, and so forth would be paid out to labor and suppliers in gold as well. The cost of mining is incurred prior to the final production of the mined gold. WIth QE, the Fed may increase its cash balance with no corresponding cost of production.
Now we have:
Several things are clear from this explanation. The first is that the initial wealth effect occurs in the Fed itself. The Fed creates this effect when it increases its own cash balances, which enables it to acquire assets which at its former—lower—level of cash balance were offered at a price that was above the Fed’s buying price. The second is that Henry has increased his net worth. Prior to the QE, he held a bond which he valued at $95 or more, which he sold for $96.
Not only is Henry’s net worth greater, but—a point which is ignored by Mr. Sproul—he has cash, which is a more marketable good than a bond. A “mark to market” net worth is not the same as cash. Ninety-six dollars in cash can be exchanged for any good that is for sale in the economy, while bonds tend to be illiquid, difficult to sell, and often can only be sold at a substantial haircut. And finally, if Jon’s reservation price for his bond is $96 or less, and if considers the Fed’s purchase of Henry’s bond as a willingness to buy similar assets for $97, his mark-to-market net worth has increased, since he now has the ability to sell a bond for $97 cash rather than $93.
During the Great Financial Crisis, the Fed purchased assets that might have been liquidated or repriced during bankruptcy reorganization at pennies on the dollar for close to par value. The markup may have been much higher than the $93 to $97 that I used in this article. In many cases, there were no buyers even close to face value. Bonds that might have been impossible to sell or could have found a buyer at a few pennies on the dollar were sold to the Fed as if they were not impaired.
Sproul’s inability to see the wealth effect comes from two errors. First, he commits a fallacy that is similar to a technique used to lie with statistics—the arbitrary choice of base year to make a percentage gain appear either larger or smaller. This is done by picking the base year before or after a large change in the most unfavorable direction to avoid that change being included in the differential. The wealth effect occurs immediately after the Fed has created money. Starting after this omits the direct cause and makes it harder to see the effect.
Sproul’s second error is ignoring the very different marketability of liquid and illiquid assets. As explained above, by creating a wealth effect for itself and then increasing its spending on assets, the Fed is putting upward pressure on mark-to-market asset values. In order to buy an asset, the Fed must offer more than the prior best bid for the asset, otherwise the owner would have already sold it.
The more assets the Fed buys, the more pronounced is the indirect effect on portfolio mark-to-market values. Without the Fed, Henry could have sold his single bond for $93, but a hundred more bondholders who also wanted to sell would have had to sell to the next marginal buyers, who necessarily are offering successively lower bids.
Sproul’s second claim, that the value of the dollar is not affected, is also clearly false. As there are more dollars and they are used to bid up the prices of bonds, the value of the dollar is diminished compared to bonds. The sellers of the bonds have increased their cash balances, and are able to bid for goods and services that they could not have bartered for with their bonds.
The effect cascades throughout the financial markets. By improving the liquidity position of the former owners of the assets, they are able to increase their spending on either consumption goods or financial assets if they wish. If the Fed continues its activities, Henry and Jon can recycle bad assets into cash and use the money to purchase more poor-quality assets that have not yet been targeted for asset purchases, on and on as long as the Fed is a willing buyer.
Contrary to Sproul, “asset swaps” in general do change wealth positions. Economist Robert Murphy makes this point with an example (either here or here). Suppose that Lockheed builds an F-35 that costs $1 billion. The Defense Department buys it from them at cost. Is that an “asset swap”? Who would claim that their wealth position was not altered by this? The plane has no other buyer than the DOD and it has no realistic market price. Murphy’s point is that a fictitious government-assigned “price” for a good that has no marketability other than to a government agency that is not constrained in its purchasing power is not the same as having that amount of cash.
Mr. Sproul attempts to show that money printing and asset purchases have no effect on wealth positions. In a single small comment he commits a number of fallacies that require a lengthy article to unpack. Even without this analysis, the claim that buying assets with money created out of nothing has no impact on wealth positions is contrary and counterintuitive—much like the claims of pop economics bloggers, which tend to be highly contrary, attention grabbing, but also false.