“Peter Cook is the author of the ‘Is That True?’ series of articles, which help explain the many statements and theories circulating in the mainstream financial media often presented as “truths.” The motives and psychology of market participants, which drives the difference between truth and partial-truth, are explored.”
Corporate stock buyback programs (CSBs) are one of two tools used by corporate CFOs who aim to “return capital to shareholders.” The other main tool is dividend policy. Dividends are prized by investors because the value of any investment is determined by the set of cash flows received over time, and dividends are the receipt of cold, hard cash.
CSBs, on the other hand, are not really a return of capital or of cold, hard cash; they are a use of a company’s capital (generated by profitable operations or by borrowing) to directly intervene into financial markets and shrink the number of shares outstanding. By shrinking the number of outstanding shares, the CSB raises earnings per share (EPS), which should theoretically raise the stock price (assuming the company earns more on its assets than its cost of borrowing).
Today, it might seem like CSBs have been around forever. In fact, they weren’t. In the aftermath of the Roaring 20s and the Crash of 1929, CSBs were seen as blatant “stock manipulation” schemes and were outlawed. It wasn’t until the 1980s that the prohibition on CSBs was seen as archaic, and they were legalized once again. The only restrictions on CSBs are that they can’t operate during the first or last half-hour of the trading day, and that they can’t operate for a few days around significant news events, such as quarterly earnings reports. By creating these restrictions, regulators are effectively admitting that CSBs affect the price of the stock they are accumulating. So, in addition to shrinking the number of shares outstanding, CSBs also push stock prices higher, but to an unknown degree.
In a normal market for any product or service, buyers attempt to buy at the lowest price and sellers attempt to sell at the highest price. In contrast, a CSB typically promises to buy a fixed number of shares during a fixed period, without regard to price. That behavior is a powerful signal to other financial market participants that the true goal of CSBs is to move the price higher. In other words, CSBs explicitly attempt to distort market prices, because they use the same method and attempt to achieve the same goal as the Fed’s QE programs (buying a fixed quantity during a fixed period without regard to price). Hence, CSBs are effectively “Corporate QE.”
Goldman Sachs estimates the flow of funds into and out of the stock market on the premise that fund flows can drive the direction of stock prices. Its most recent chart of estimated flows is shown below.
Since 2010, CSBs have consistently been the largest flow into or out of the US stock market. Also, they don’t behave as other market participants because other buyers want to pay the lowest price, not just any price. So, considering the size and intensity of their buying, CSBs have more influence on stock prices than pension funds, ETF investors, foreign investors, insurance companies, mutual funds, and individual investors. In fact, in many years CSBs alone are larger than the activities of those investor types combined. For example, since 2010 CSBs have accounted for somewhere between 120% and 300% of total net stock market annual inflows. It is doubtful that regulators anticipated that result when they legalized CSBs in the 1980s.
CSBs are also notable for their consistency in both direction and amount. Most of the other types of stock market investors may buy one year and sell another year. Or, in the case of pension funds, they may be consistent sellers to fund retiree benefit payments, but the amount of selling varies annually.
In summary, looking at the activity of all stock market investors, CSBs are most probably the largest determinant of stock prices. Put another way, corporate management teams, whose compensation is significantly based on lifting the stock price, execute CSBs, which are the largest determinant of stock prices. This combination of compensation arrangement and responsibility for executing CSBs opens the potential for a “principal-agency” conflict of interest. If corporate managers (agents) are skilled at buying stock on behalf of shareholders (principal), then CSBs may not produce a conflict of interest. But as we will see, the data actually shows a lack of skill by corporate management at executing CSBs.
The same type of table from Goldman Sachs is shown below, but this one was produced in mid-2016 and stretched back to 2008. The data on corporate behavior is near the bottom, highlighted by the red oval. Focusing on 2008 and 2009, a new pattern of CSB behavior emerges.
Instead of being buyers, corporations were net sellers during 2008 and 2009. How could that be? Think of CSBs as a discretionary purchase, just like a decision made by any purchaser. When your cash flow is high, you can afford to buy things. When your cash flow is low, you don’t buy discretionary items. In fact, you might be forced to sell previous purchases or even borrow money to meet current expenses.
So too it is with corporations. Their CSBs are discretionary. Corporations buy stock when cash flow is high, and they are forced to sell stock when cash flow is poor. Because cash flows are correlated with stock prices, CSBs buy stock when prices are high and sell stock when prices are low. In other words:
- The level of CSB purchases are dependent on fluctuations in the business cycle, and
- Corporations tend to buy high and sell low, so their skill at market-timing is poor.
While these might seem to be shocking statements, additional historical data in the chart below validates them.
Source: Societe Generale
Note: The chart ends in 2015, but separate data show a recent plateau in both buybacks and debt issuance
The chart above shows “net” buybacks (blue line), so it differs slightly in magnitude from the Goldman data, but it shows the same directional pattern of rising and falling with the business cycle. In addition, this chart includes data on the change in corporate debt (red line). Thus, we can draw the following conclusions:
- In the 1990s, the level of CSBs and debt issuance was lower, and they did not rise and fall at the same time.
- Since 2000, CSBs show a clear pattern of increased buying during business expansions and a decrease during recessions, which occurred in 2001 and again in 2008-09. Because stock prices are higher during expansions and lower during recessions, CSBs indeed appear to be poor market timers.
- Corporations increase their debt during business expansions and pay down debt during recessions. Because interest rates are higher during expansions and lower during recessions, corporations also appear to be poor market timers when it comes to the issuance of debt.
- Given the similarity in magnitude and direction of the blue and red lines, many corporations borrow money to fund CSBs. While the Fed’s low-interest rate policy reduced the cost of borrowing post-2009, borrowing money to fund CSBs also occurred prior to 2009, so factors other than Fed policy are relevant.
- The fluctuations in CSBs and debt are becoming greater over time, meaning that corporations are doing more of what they are poor at doing.
The preceding charts demonstrate the “big picture” analysis of aggregate CSBs. But specific sectors may experience recession even while the economy is growing. After the crash of oil prices from $100 to $26 that occurred between November 2014 and February 2016, energy companies were forced to sell stock at half the price that they had purchased in CSBs before November 2014 (e.g., Hess Corp.). Similarly, several years ago retailers used CSBs to purchase stock at much higher prices than today’s prevailing prices. Losses can run into the billions of dollars for individual companies (e.g., Macy’s Inc.). In each case, instead of diverting cash flow into financial engineering tactics such as CSBs, management could have been investing cash flow to improving their operating businesses.
QE, as practiced by the Fed, is the policy of issuing currency out of thin air to buy government bonds and mortgage-backed securities. Globally, central banks issue currency out of thin air to buy a variety of other securities, including corporate bonds, junk bonds, and stocks. Many central banks, including the Fed, directly intervene in financial markets buy a specific quantity of securities during a specific period, and with no price limit. Regardless of what is purchased, the goal is to push the price higher and yields lower, because central banks have stated their beliefs that high financial prices give people confidence to spend more than they otherwise would, increasing GDP growth in the process. This is known as the Wealth Effect.
Corporate QE, as practiced by corporations executing CSBs, means issuing bonds at low-interest rates or using cash reserves to purchase stock in financial markets. So Corporate QE has an opportunity cost that central bank QE doesn’t. But the effect of Corporate QE on market prices is similar to QE practiced by central banks. CSBs directly intervene in stock markets to buy a specific amount of stock during a specific period, and with no price limit. Because CSBs are the largest inflow to the stock market, they probably succeed in push stock prices higher, treating investors to gains they wouldn’t otherwise have. During this part of the cycle, CSBs seem like a great idea, and the principal-agency problem is not visible.
But that’s only half the story. When the business cycle rolls over, cash flows decline and CSBs are suspended. If things get really bad, the data from the past 20 years shows that corporations are forced to issue stock at much lower prices than the prices paid by CSBs. Unfortunately, the goal of “returning cash to shareholders” via morphs into borrowing or issuing equity to survive. During this part of the cycle, previously-executed CSBs don’t seem like a very good idea, and the principal-agency problem is highly visible.
The recent tax reform legislation included a provision for companies to repatriate cash held overseas. There is some debate as to whether the repatriated cash will be used to fund new CSBs, but that’s what happened in a similar change in repatriation rules in 2005. Some estimate that hundreds of billions of dollars will be repatriated in coming months/years to fund CSBs.
If so, incremental Corporate QE will soon be in full swing to counteract the unwind of QE by the Fed and the upcoming tapering of QE by the ECB and BOJ, at a time when even Janet Yellen admits that stock valuations “are on the high side.” Confirming this suspicion, Goldman Sachs recently stated that, during the market swoon in early February, the GS buyback desk had its “most active week in history.” Corporations are “buying high” once again, just as they did ten years ago, and Corporate QE was at least partially responsible for producing the stock market’s best weekly gain in seven years.
CSBs, unlike QE, incur an initial cost, because corporations fund CSBs by issuing debt or with cash from operations that could have been invested elsewhere. But CSBs and QE share a method of price-insensitive buying that is unlike the behavior of any other market-based buyer; they buy a fixed amount during a fixed period, with the goal of buying at ever-higher prices.
In analyzing the success of CSBs, or for that matter, QE, one should wait until a full cycle before declaring a judgment. Unfortunately, the data shows that CSBs, in aggregate and over time, buy high and sell low. When cash flow and share prices are high, CSB buying is at its highest. When cash flow and share prices are low, CSB buying is reduced or even reversed by share issuance. Because the fluctuations of CSBs and debt issuance have increased over the past 20 years, corporations are doing more and more of something they do poorly. Over an entire business cycle, CSBs may destroy value, not create it.
If so, one explanation is the principal-agency problem, which occurs when the agent has different incentives than the principal. Among large companies, it is common for a significant portion of executive compensation (stock options, incentive rewards, etc.) to be tied to the stock price. If the tenure of senior corporate management (agent) is shorter than a full business cycle and its ups and downs, the principal-agency problem for the shareholder (principal) will tend to be greater.