Master’s Thesis
M.Cs. Economic Development & Globalisation
BUYBACKS: HAS MAXIMIZING SHAREHOLDER
VALUE GONE TOO FAR?
The effect of buybacks on investment
Tom van der Zanden
Abstract
Theoretical literature seems to agree that stock market development is good for economic development. That is why it is interesting to see that share repurchases (i.e. buybacks) are issued more and more, since they decrease stock market volume per definition. This paper will further investigate this relationship between share repurchases and economic growth. In particular, it will examine the effect of buybacks on investments. The main findings are that investments do seem to be negatively affected by buybacks, but this disinvestment is mostly in intangible assets which makes it hard to conclude what effect this has on the company.
Key words: share repurchases, investments, capital expenditures
JEL classifications: D25, G11, G31, O16
Acknowledgements
I would like to thank prof. dr. Bezemer for his feedback and guidance throughout the entire process of this thesis. Even through these unprecedented times of a global pandemic which caused some trouble along the way, he made sure that I kept improving and thinking critically on my own work.
Table of Contents
1. Introduction 3
2. Theoretical background 6
3. Methodology 10
3.1. The main variables 10
3.2. The sample 12
3.3. A first exploration of the data 12
4. Empirical Results 16
4.1. The Capex model 16
4.2. The asset-based models 17
4.3. Robustness check: heteroscedasticity 18
4.4. Discussion 19
5. Conclusion 20
Appendix A: Definition of Variables 22
Appendix B: S&P 100 companies 23
Appendix C: Buybacks over time 24
Appendix D: Collinearity Matrix 26
1. Introduction
There is a long-standing literature on financial development and economic development, in which the distinction between market-based finance (stocks and bonds) on one hand and bank loans on the other has played a large role (Feder, 1980) (King & Levine, 1993) (Shan et al, 2001).
This paper will be about the question if stock markets support investment as money flows through stock markets from households to firms. I will take into account changes in stock markets over the last decades which make them different from textbook explanations. In particular, in the last decades stock repurchase programmes have grown tremendously in size; “from 2000 through 2015 298 companies expended 945 billion Euro on stock repurchases.” (Sakinc, 2017) Figure 1 visualises the increase in share repurchases over the last two decades, albeit with a fall in the economic crisis of 2008. In fact, the rise of buybacks is so strong that since 1999 the general trend is that there are more stocks repurchased than issued in the S&P 500 , as figure 2 shows. The red line in figure 2 shows the net issuance of non-financial corporate
1
businesses, which compared to the blue line (which is total net issuance) is much less volatile, showing that financial companies more aggressively engage in issuance or repurchasing of stocks. The increasing trend of share repurchasing means that money is flowing from firms to households, which seems counterintuitive for economic development. There are economists however, that see stock repurchases (and dividends for that matter) as a reallocation of capital from mature to young companies that are in dire need of a capital-injection. (Edmans, 2017) However, there is also a body of literature that views share repurchases as a value extracting mechanism, mainly used for market manipulation. (Lazonick, 2015) (Sakinc, 2017) Is the relation between the growth of stock markets and the growth of investment changed by stock repurchase programmes?
In order to investigate this I first establish through theoretical literature that stock market development is good for economic development. Harris (1997), Choong et al (2010), and Van Nieuwerburgh et al. (2006) provide evidence that for both developing and developed countries stock market development has a positive effect on economic growth. Thus I raise questions on why share repurchases become more and more popular, as they lower stock market volume per definition.
1A market-capitalization-weighted index of the 500 largest U.S. publicly traded companies. The index is widely
share repurchases is one of self-interest. Managers themselves are through their compensation often shareholders of the company they work at, hence by boosting the share price and earnings-per-share by buying their company’s own share, they boost their personal wealth. Given these theories and justifications it remains that share repurchases are of value extracting nature. Lazonick (2014) makes the case that all the capital spend on buybacks could have been spent on investments, R&D or job creation. Grullon & Michaely (2004) actually find evidence that companies lower their investments after they repurchase shares. However, their data is from a period where buybacks were much less prominent. Hence, this paper adds to the literature additional research whether share repurchases impact economic growth, through lowering investments.
To assess this relation I conduct three different regressions where capital expenditures, change in total assets, and change in non-current assets are the dependent variable respectively. These models are in line with the literature from Harford & Kolasinski (2014) and Asker et al. (2011). The sample of companies used are taken from the S&P 100, which is a subset of the aforementioned S&P 500. Hence I have companies that are amongst the most traded and liquid stocks, and are thus likely to be prone to stock repurchases in order to boost the stock price or EPS.
The main findings from these regression analyses is that for capital expenditures as a proxy for investment, I could find a significant relationship regarding share repurchases. However, the first version of the model actually gave a positive effect between the two, but this result is inconclusive, as the lagged version was not significant by a fair margin. For the other two asset based models, I did find a significant negative relationship between share repurchases and investments, even after a robustness check. Hence, this paper provides some evidence that with share repurchases, investments go down. However this disinvestment originates mainly from intangible long-term assets given that no negative relationship was found for capital expenditures which are investments in tangible assets. It is hard to say what effect this disinvestment in intangible assets has on the firm, as intangible assets have much less impact on productivity than tangible assets, but do provide financial stability.
2. Theoretical background
In order to understand the impact of buybacks on economic development, an understanding of the effect of stock markets on economic development must be established first. In Stock Markets
and Development (1993) Atje and Jovanovic established that there is a large effect of stock markets on subsequent economic development. In a much cited rebuttal, Harris (1997) argued that this effect is probably much weaker than Atje and Jovanovic claimed. This, as Atje and Jovanovic used a lagged variable for investment and stock market activity in order to tackle endogeneity issues. However, Harris argues that this is not an adequate solution since the lagged variables are “not highly correlated with current investment and hence not a good proxy for this variable.” (Harris, 1997). This gives rise to omitted variable issues, and so Harris argues that Atjes and Jovanovic their coefficient is biased upwards. After a revised 2SLS model this is indeed confirmed. Besides, he differentiated the effects for developing and developed countries, with the notion that for less developed countries “the stock market effect, (...) is at best very weak. For the developed countries, however, stock market activity does have some explanatory power.” (Harris 1997). In line with this distinction, some literature has specified stock market activity effects on economic growth in either developing or developed countries.
Choong et al. (2010) re-establishes the importance of stock markets for developing countries. In their research of 32 developing and 19 developed countries over the period of 1988 to 2002 they found that “the negative impact of private capital flows can be transformed into the positive if the stock market development has reached a certain minimum (threshold) level,” and “that domestic stock markets play a pivotal role in transforming the negative impact of private capital flows on economic growth.” (Choong et al, 2010) In other words, private capital inflows that have a negative impact on economic growth in developing countries (debt, portfolio investment, and in some cases FDI) can turn into having a positive impact once a certain level of stock market development has been achieved. Thus, once a threshold level in stock market development has been reached, the effect of stock market development on economic growth turns positive.
specifically focusing on stock market development, they find “an important long-term relationship between stock market development and economic growth in Belgium, especially in the period of rapid industrialization [1873-1935].” (Van Nieuwerburgh et al, 2006) However, for the period 1917-2002 they too find that “Stock market development is a driving force of economic development in this period.”
All in all, there seems to be a consensus amongst the literature that there is, even though the weight is disputed, a positive effect of stock markets on economic activity. That is why it is interesting to see that stock repurchases (i.e. buybacks), have become increasingly popular amongst big firms over the last two decades, as Lazonick (2015) and Sakinc (2017) show. Buybacks are namely directly decreasing stock market volume by extracting the number of shares outstanding in the stock market. Though, it must be noted that stock market capitalization (stock price times volume) could increase, which is dependent on the effect of buybacks on the stock price.
Before I dive into the capital extracting nature of buybacks, a technical clarification is at hand. There are namely different types of buybacks, of which the two major ones are tender offers, and open market repurchases. The former is a construct where a company offers shareholders directly a predetermined price, and those shareholders who agree tender their shares to the company. This type of stock repurchase is usually done by a company in order to regain ownership of their own company, and relieve (short-term) pressure from shareholders. However, “tender offers constitute only a small portion of modern buybacks.” (Lazonick 2014) The other type, open market repurchases, are much more common. These repurchases, as the name suggests, are the company simply buying its own shares on the open stock market. Even though legislation differs across countries, it is basically a legal form of stock market manipulation (Lazonick 2014). This is the form of buybacks that this paper will focus on.
Yet, if buybacks are so value extracting, why are the majority of companies still doing it, and in an increasing manner as well? Abraham et al. (2018) argue that share repurchases play a major role in keeping investor confidence up. There are several theories that try to explain this hypothesis. Nohel & Tarhan (1998) summarize the two most widely accepted. The first one is the information signaling hypothesis, which states that “a company's willingness to pay a premium to purchase its own shares sends a strong signal to lesser-informed outside investors that the company's future prospects are improving.” (Nohel & Tarhan, 1998). The second theory is the free cash flow theory, which states that “firms with excess cash and a poor portfolio of investment opportunities will face sizable agency costs if the excess cash is not distributed to shareholders.” (Nohel & Tarhan, 1998) However Nohel & Tarhan also argue that companies with poor asset portfolios use this argument to sell off poor performing assets and distribute the proceeds of the sale as share repurchases. This will boost performance in the short run (as the poor performing asset has been sold), but could hurt the company in the long run, as the investment value is extracted from the company.
A third theory of why share repurchases are used to increase investor confidence is that of dividend substitution. Abraham et al. (2018) argue that dividends used to establish investor confidence in the same way share repurchases do through the information signalling and free cash flow theories. However, since all managers want to signal that the company is doing well and financially sound, dividends have become an essential part of “proving” that, to the point where companies are even willing to go in debt just in order to keep paying the usual dividends. “Since it is the norm to pay out some level of dividends, the only incremental information conveyed by dividends is that their reduction indicates the firm’s inability to achieve a minimum level of earnings.” (Abraham et al, 2018) Thus, since share repurchases are much more volatile, the dividend substitution theory states that share repurchases are a much more accurate representation of manager confidence in future earnings.
repurchases from a managers perspective. So even though it might be a positive side-effect of share repurchasing, it is not a reason for managers to engage in them.
The aforementioned theories regarding investor confidence provide some answers on why managers increasingly engage in the act of share repurchases. However, these seemingly positive effects are still of value extracting nature, as funds that could have gone to capital expenditures or employee compensation have gone to outside benefactors. I define value extraction here as capital leaving the company for little to nothing in return. Sure, the company regains control of a small portion of its shares, but this does not benefit the company and it is not the goal of the company either. Besides, Abraham et al (2018) argues that with the internet, there is such a vast amount of information about companies available, including their assets, opportunities, image, etc., that nowadays, more than ever “firm value is viewed in terms of the earning power of assets, not the signaling power of market imperfections, such as dividends and shares repurchased.” (Abraham et al, 2018) Thus, investor confidence will be best influenced if the company will use the funds to invest in projects with a positive NPV, instead of share repurchases.
A final explanation for the increase in buybacks of the last two decades is one of self interest. A large share of managers’ compensation comprises stock-based instruments, which in the US they are allowed to sell as they please. “In 2012 the 500 highest-paid executives named in proxy statements of U.S. public companies received, on average, $30.3 million each; 42% of their compensation came from stock options and 41% from stock awards.” (Lazonick, 2014) Hence, by engaging in share repurchases, they increase the value of their personal wealth by pushing up the share price of the stocks they receive. Kahle (2002) finds that indeed, managers compensation is a strong motivation for a company to repurchase shares.
In this literature review I have already established that through lowering stock market volume, buybacks have a negative impact on economic growth, even though there might be some positive side effects. This paper will add to this literature the negative impact of buybacks on investments, which indirectly hinders economic growth as they are “trillions of dollars that could have been spent on innovation and job creation in the U.S. economy over the past three decades [that] have instead been used to buy back shares.” (Lazonick, 2014)
In addition, I expect his negative relation since corporate investments have two main sources of financing: debt or retained earnings. The latter should be the prioritized choice, since there is no payback or interest payments associated with them, as is with the prior. Hence, investments should originate from retained earnings. However, both dividend payments and stock repurchases are financed from debt and retained earnings too. Hence, with an increase of stock repurchases, I expect investments will go down (given that dividend payments stay constant, or debt goes up). Grullon & Michaely (2004) provide evidence for this, as they find “that repurchasing firms decrease their investments” in the years after the repurchase. However, this paper used a dataset which ranged from 1980 to 1997. Share repurchases in this period were much smaller, as total repurchase value over the sample period was 427.6 billion US dollar, while this value is matched in just two quarters in 2019, as figure 1 shows.
Thus this paper will add to the literature a more recent analysis of the effect of buybacks on economic growth, in a time period where buybacks are much more prominent. In particular, I will empirically research the effect between buybacks and corporate investment, which I expect to be negative. Thus the question this paper will try to answer is do buybacks have a negative impact on investments? From which the following hypothesis is formulated:
3. Methodology
In order to investigate the relationship between share repurchases and investments, this paper uses an ordinary least squared (OLS) regression model with fixed effects. I use this model, as I expect investment levels to differ across companies, but the effect of buybacks on investment to be constant across companies. This paper makes use of data provided by stockrow.com, a public website that collects data on publicly traded American companies, including balance sheets, cash flow statements and other financial metrics. Both the data for the dependent as the independent variables are gathered from this website. The time range on all variables is 30-06-2010 to 31-03-2020.
3.1. The main variables
The main explanatory variable is share repurchases. This is directly provided by stockrow.com as net common equity repurchases. There could be some concerns on the validity of this variable given that it is ‘net’ repurchases, which means that new issues of equity are pre-calculated into the variable. However, for this research this does not pose a problem, given that new issues of shares (which lowers net repurchases) should increase investment, which would give the same effect as lowering net repurchases in the first place. Admittedly, not all capital raised from new public offerings will always go towards investments, but it is an assumption I make for simplicity’s sake. In order to deal with simultaneity issues this variable will be transformed in a lagged variable of t-1. This, as I want to research the effect of the decrease in retained earnings due to share repurchases on investments in the following fiscal period, which in this dataset will be a quarterly year. However, since if you start planning to repurchase shares one might already lower investments which would result in a reduction in investments in the same time period. Therefore both models with t and t-1 will be estimated.
Other papers use the change in assets of the company; in their working paper, Asker et al (2011) use gross investment, defined as “the annual increase in gross fixed assets scaled by beginning-of-year total assets.” Another measurement of investment is capital expenditures, which is defined as the change in property, plant, and equipment plus current depreciation. This is in line with similar literature like Markarian & Michenaud (2019). This paper will use both measurements in two separate models in order to be most complete and to see if there are differences in the type of investment changes. The slight difference to my paper is that for the Asker et al. definition, quarterly change is used, instead of annual change.
For the remainder of the model, this paper lends heavily from Harford & Kolasinski (2014). In their second equation they test for investment levels pre- and post buyout, with an ordinary least squared regression (OLS) with fixed effects. To be precise, they “examine the sensitivity of their investment to cash flow pre- and postbuyout while controlling for firm- and industry-level investment opportunities.” (Harford & Kolanski, 2014) This regression analysis is based on the traditional investment-cash-flow models introduced by Fazarri et al. (1988). Although it must be noted that these models are constructed around the investment-cash-flow sensitivity analysis, they also measure the effect of investment opportunities on investment levels. In addition, the main variable regarding cash flows in these models (EBITDA/assets) happens to be a control variable for investment opportunities as well. Hence, while excluding the variables concerning buyouts, and including share repurchases as the main independent variable, a literature tested model for investment is provided.
Other models were looked at to add to my model as well. Grullon & Michaely (2004) use a model that measures changes in capital expenditures from t-2 to t+3 where shares are repurchased at t. However, since in my sample shares are repurchased on almost a continuous basis (see below and Appendix B) this model did not fit my sample. Most other literature use investments as independent variable, like Koo & Maeng (2019), Boudry et al. (2013), and Asker et al. (2011). This makes most of the control variables not viable for our model. Control variables from these other models that could potentially be viable like firm size or stock capitalization are irrelevant due to the sample (which only entails large established companies, see below) and are therefore also not used.
have no influence on the outcomes. Finally, Harford and Kolanski use two interaction variables which are irrelevant for my hypothesis and are therefore not used. Appendix A shows the definitions of all variables mentioned. In conclusion, the first statistical model is as follows:
assets cap. ex.
i,t = α1, i+ β1buybacks i, t+ β2salesgrowth i, t + β3EBIT DAassets i, t + ε
(1)
In which denotes company, denotes time, and i t α is the constant with i = 1due to the fixed effects model. The second model, with percentage asset growth as its proxy for investment is as follows:
3 ΔAssetsi, t = α1, i+ β1buybacksi, t+ β2salesgrowthi, t+ β EBIT DAassets
i, t+ ε (2)
This second model has the theoretical downside that buybacks are usually financed by assets, unless financed by debt, but as discussed in the literature review debt financing should not be the preferred method. Therefore it could be argued that the negative relation between buybacks and assets is fundamental, as automatically with an increase in buybacks, assets should go down (due to the purchases of shares). Hence, any result on the relationship between buybacks and investments could be interpreted as insignificant. Therefore I propose a third model, which uses percentage change in non-current assets as a proxy for investments. This model would show if there is a significant relationship between buybacks and non-current assets, which if proven significant, would mean that long-term assets are affected by buybacks, and is hence an indication of a company’s long term (dis-)investment. Thus:
3 ΔNon− current assetsi, t = α1, i+ β1buybacksi, t+ β2salesgrowthi, t + β EBIT DAassets
i, t+ ε (3)
3.2. The sample
As a company sample I use the S&P 100 components, e.g. the 100 companies that make up the shared index. The reason for this is as “the S&P 100, a subset of the S&P 500®, is designed to measure the performance of large-cap companies in the United States. The index comprises 100 2 major blue chip companies across multiple industry groups3 .” (us.spindices.com). Thus, these
companies are amongst the most traded and liquid stocks, and are thus likely to be prone to stock repurchases in order to boost the stock price or EPS. Since the components of the S&P 100 change over time, I decided to take the components as of the date the data was gathered (for a full list see appendix B). To clarify, I take the historical data of the companies that are in the S&P 100 at 06-05-2020, and do not change the sample companies in that time period to be identical to the S&P 100 (which changes its components over time). This, as I am not necessarily
interested in buybacks within the S&P 100, but of a sample of companies with a strong relation to the stock market (e.g. are traded a lot, are prone to repurchases). The S&P 100 provides quite adequately, given the market capitalization and financial/economical significance of its components. This way, the sample will be balanced, as the companies stay fixed during the sampled time period. Ironically, the S&P 100 currently has 101 companies listed, due to Alphabet Inc. having two classes of stock. However, since these are representing the same company, only one will be considered in the estimations.
I am aware that since these are all blue-chip companies, there is a bias towards mature companies. This gives more merit to the free cash flow and reallocation of capital theory. These companies thus could have more incentive to repurchase shares than younger companies listed on the New York Stock Exchange. However, this does not take away that I want to test if investments go down in these companies after share repurchases.
3.3. A first exploration of the data
Before I give a first overview of the data, I must mention two small data transformations I have computed for readability’s sake. First, since stockrow.com provided equity repurchases and capital expenditures as cash flows, they were presented in a negative value. Hence, I reversed the sign of both variables so that an increase in buybacks or capital expenditures leads to a higher value for the variable buybacks and Capex. Second, I scaled Buybacks by one million, for the sole purpose of readability. All following computations that variable are therefore in million US dollar from this point forward. The growth variables are percentage based, e.g. a value of 0.01 means 1% growth.
Regarding outliers, for buybacks I deleted two, which were caused by an error of a company which issued 46 billion of shares only to take them back the next quarter, skewing the data. For Sales Growth I deleted observations of over 1000% growth or decline, as they are probably errors in the data, as such growth rates seem very unrealistic for established companies in the S&P 100. Table 1 below shows the descriptive statistics for the main variables, corrected for these outliers.
Table 1: Descriptive statistics
The difference between companies is also interesting to notice. There are three companies that stand out by glancing over Appendix B: Amazon, Kraft Heinz, and Apple. Amazon and Kraft Heinz stand out because they are the only two companies in the dataset that do not systematically repurchase their own shares. The difference between the two is that Kraft Heinz does systematically issue dividends, while Amazon doesn't do that either. Amazon is thus the only company in the dataset that does not provide any direct capital to their shareholders.
The other extreme is Apple Inc. This company repurchases the most shares in the dataset by a fair margin. In the final quarter of 2019, this company alone repurchased shares for 20 billion US dollars. If I compare this to the total value of buybacks of the S&P 500 from figure 1, it shows that this single company is responsible for roughly 10% of the buybacks in this period. It is interesting to see how two such large and influential companies have such a different approach towards share repurchasing. There is no need to delete these observations as Amazon and Kraft Heinz will not be taken into the regressions, and Apple does not seem to alter the trend as seen in figure 3. Figure three also shows a confirmation of what the introduction and literature review stated, that there is a trend of an increase in share repurchasing in the last decade.
Finally, to address concerns for collinearity the correlation matrix of Appendix D was produced. Since there are no strong correlations between independent variables there are no concerns for collinearity, and the regression analyses as stated can be performed.
Figure 3: Total Buybacks (in Million USD) of sample, with (upper) and without (lower) Apple Inc.
Model estimator Dependent variable Chi2 Prob>Chi2
(1) Capex 1.20 0.5497
(2) Asset growth 45.57 0.0000
(3) NC-Asset growth 38.52 0.0000
4. Empirical Results
In this section I will conduct the regression analyses specified in the previous chapter. First, I will estimate and discuss them separately, afterwards I will discuss what the differences in results means for our hypothesis.
4.1. The Capex model
4.2. The asset-based models
In order to further investigate the relationship between share repurchases and investments I conduct the second and third regression analysis based on the investment proxy of Asker et al., namely the percentage change in gross assets, and this paper's specification of the percentage change in non-current assets. The results are shown in table 3.
4.3. Robustness check: heteroscedasticity
Finally, I address concerns for heteroscedasticity. Theoretically, larger companies, e.g. companies with more assets, have relatively more funds to repurchase shares. Therefore, one would expect a natural correlation between assets and buybacks. There are two reasons why this does not invalidate my results. Firstly, since my models use the percentage change in assets, the change in assets should be relatively equally impactful amongst companies. Secondly, after correlating total assets and buybacks the relation was to be called weak at best (0.2165). Finally, after a visual check on correlation matrices, the result was inconclusive and thus I decided to run the regression with robust standard errors, for completeness sake. The results of these regressions can be found in table 4 below.
4.4. Discussion
Ideally, to support the hypothesis that higher values of share repurchases would lead to a decrease in investments, the coefficients of Buybacks would have been negative and significant in all three models and in all versions. However, this is not the case. In my dataset I could not prove a significant negative relationship between buybacks and capital expenditures (model 1). In fact, for the version without lagged buybacks the sign was positive and significant. However, given that the lagged buybacks version was not significant it is hard to draw robust conclusions from this model. It could be evidence towards the free cash flow theory, as when there is excess cash it will be both distributed to share repurchasing and capital expenditures, but again, this is inconclusive. In general, it means that I have no evidence that buybacks negatively influence investments in property, plant, and equipment, e.g. capital expenditures.
However, all coefficients for the asset-based models do prove to be negative and significant (model 2 and 3). This is the case for both the standard and the lagged version of the models. This provides further evidence for what Grullon & Michaely (2004) stated, that repurchasing firms reduce their investments. Interestingly, their main proxy for investment was capital expenditures, for which I could not prove a negative relation. That does not give away that assets and noncurrent-assets do seem to be significantly impacted by share repurchases. Hence, these models provide evidence that buybacks negatively influence a more general definition of investment. This could imply that, given that capital expenditures did not prove significant, companies use long term intangible assets in order to finance their share repurchases. This should not hurt the productivity of the firm (as that is generally generated by tangible assets), but it can hurt the profitability and financial position of a firm.Therefore share repurchases could negatively impact the firm in general. This effect is, albeit weaker, still robust after adjusting for any concerns for heteroscedasticity.
5. Conclusion
This paper investigates the relationship between stock market development and economic growth. In particular, it focuses on the relation between share repurchases and investments. After a sharp decline in the economic crisis of 2009, share repurchases have increased fourfold. In the literature there is disagreement whether this is a positive or a negative development.
The main arguments in favour are first of all the free cash flow hypothesis, which states that “firms with excess cash and a poor portfolio of investment opportunities will face sizable agency costs if the excess cash is not distributed to shareholders.” (Nohel & Tarhan, 1998) Second, both the signaling theory and the dividend substitution theory state that share repurchases are the best way to keep investor confidence up. A final argument in favour is that share repurchases are, just like dividends, an excellent way to distribute capital form mature companies with a lot of capital and little investment opportunities to young companies with many investment opportunities but a severe lack of capital.
There are however, also many arguments against share repurchases. William Lazonick is one of the main authors of the opposition. He argues that the main justifications of buybacks, namely confidence signaling, offsetting EPS dilution, and the free cash flow theory are all either ungrounded or a failure of management. Lazonick and Sakinc both see share repurchases as “the value extraction orientation of big corporations that is largely shaped by the maximization of shareholder value ideology.” (Sakinc, 2017) Grullon & Michaely (2004) find evidence that after the repurchasing of shares, capital expenditures in the following years are lowered. However, this was in a time where share repurchases were much less voluminous and prominent. Therefore I conducted research to see if in the last decade share repurchases still have a negative impact on investments.
The main findings of this paper are that there is some evidence that share repurchases lower investments, but it is inconclusive what effect this has on the company. This, as the first model did not show a negative sign on capital expenditures and one of the versions was significant, which would disprove the hypothesis. However, the asset based models did all show a negative significant sign, which would confirm the hypothesis. After a robustness check for heteroscedasticity this result still holds. Therefore I came to the conclusion that firms probably use intangible long-term assets to partly finance their share repurchases. However, whether this is plain disinvestment or a restructuring of the asset portfolio of a firm is unknown.
limits the external validity of this paper, as the result might have been different amongst younger, less-established companies, or very mature companies with a lower market capitalization. A second limitation is that the dependent variable was not reported as is, but as net share repurchases. With theoretical reasoning I argued that this was not a problem, but some would argue that this assumption does not hold in reality, as not all proceeds of share issuance go towards investments.
Appendix A: Definition of Variables
Variable Label Definition
Capital expenditures Capex The change in property,
plant, and equipment plus current depreciation
Sc_Capex (scaled capital expenditures)
Capex scaled by Total Assets
Total Assets Assets The total of assets on the
balance sheet of a company.
Asset Growth Assetgrowth The percentage change in
total assets from t-1 to t
Non-current Assets NCAssets The total of all non-current or 4
long-term assets on the balance sheet of a company
Buybacks Buybacks The total value of shares
repurchased minus the total value of shares issued (e.g. net repurchases), in million USD
lnBuybacks The total value of net
repurchases at t-1, in million USD
Sales Growth Salesgrowth The percentage change of
revenue from t-1 to t
EBITDA EBITDA Earnings before interest,
taxes, depreciation, and amortization
Sc_EBITDA (scaled EBITDA) EBITDA scaled by Total Assets
All variables are denominated in US Dollars
Appendix B: S&P 100 companies
5Apple Inc. Capital One Financial Corp. International Business Machines
Occidental Petroleum Corp.
AbbVie Inc. ConocoPhillips Intel Corp. PepsiCo
Abbott Laboratories Costco Wholesale Corp. Johnson & Johnson Pfizer Inc
Accenture Cisco Systems JPMorgan Chase & Co. Procter & Gamble Co
Adobe Inc. CVS Health Kraft Heinz Philip Morris International
Allergan Chevron Corporation Kinder Morgan PayPal Holdings
American International Group DuPont de Nemours Inc The Coca-Cola Company Qualcomm Inc.
Allstate Danaher Corporation Eli Lilly and Company Raytheon Technologies
Amgen Inc. The Walt Disney Company Lockheed Martin Starbucks Corp.
American Tower Dow Inc. Lowe's Schlumberger
Amazon.com Duke Energy MasterCard Inc Southern Company
American Express Emerson Electric Co. McDonald's Corp Simon Property Group, Inc.
Boeing Co. Exelon Mondelēz International AT&T Inc
Bank of America Corp Ford Motor Company Medtronic plc Target Corporation
Biogen Facebook, Inc. MetLife Inc. Thermo Fisher Scientific
The Bank of New York Mellon FedEx 3M Company Texas Instruments
Booking Holdings General Dynamics Altria Group UnitedHealth Group
BlackRock Inc General Electric Merck & Co. Union Pacific Corporation
Bristol-Myers Squibb Gilead Sciences Morgan Stanley United Parcel Service
Berkshire Hathaway General Motors Microsoft U.S. Bancorp
Citigroup Inc Alphabet Inc. (Class C) NextEra Energy Visa Inc.
Caterpillar Inc. Alphabet Inc. (Class A) Netflix Verizon Communications
Charter Communications Goldman Sachs Nike, Inc. Walgreens Boots Alliance
Colgate-Palmolive Home Depot NVIDIA Corp. Wells Fargo
Comcast Corp. Honeywell Oracle Corporation Walmart
Exxon Mobil Corp.
Appendix D: Collinearity Matrix
SC_Capex Assetgrowth NCAssetgrowth Buybacks Sc_EBITDA Salesgrowth
References
Abraham, R., Harris, J., & Auerbach, J. (2018). Determinants and Consequences of Share Repurchase Decisions. Modern Economy, 9(05), 966.
Antonios, A. (2010). Stock market and economic growth: an empirical analysis for Germany. Business and Economics Journal, 2010, 1-12.
Atje, R., & Jovanovic, B. (1993). Stock markets and development. European Economic Review, 37(2-3), 632-640.
Asker, J., Farre-Mensa, J., & Ljungqvist, A. (2011). Comparing the investment behavior of public and private firms (No. w17394). National Bureau of Economic Research.
Baker, H. K., Powell, G. E., & Veit, E. T. (2003). Why companies use open-market repurchases: A managerial perspective. The Quarterly Review of Economics and Finance, 43(3), 483-504.
Boudry, W. I., Kallberg, J. G., & Liu, C. H. (2013). Investment opportunities and share repurchases. Journal of Corporate Finance, 23, 23-38.
Choong, C. K., Baharumshah, A. Z., Yusop, Z., & Habibullah, M. S. (2010). Private capital flows, stock market and economic growth in developed and developing countries: A comparative analysis. Japan and the World Economy, 22(2), 107-117.
Chung, K. H., & Pruitt, S. W. (1994). A simple approximation of Tobin's q. Financial management, 70-74. Edmans, A. (2017). The Case for Stock Buybacks. Harvard Business Review,
https://hbr.org/2017/09/the-case-for-stock-buybacks
Fazzari, S., Hubbard, R. G., & Petersen, B. C. (1987). Financing constraints and corporate investment (No. w2387). National Bureau of Economic Research.
Feder, G. (1980). Economic growth, foreign loans and debt servicing capacity of developing countries. The Journal of Development Studies, 16(3), 352-368.
Grullon, G., & Michaely, R. (2004). The information content of share repurchase programs. The Journal of Finance, 59(2), 651-680.
Harford, J., & Kolasinski, A. (2014). Do private equity returns result from wealth transfers and
short-termism? Evidence from a comprehensive sample of large buyouts. Management Science, 60(4), 888-902.
Hsieh, C. T., & Klenow, P. J. (2010). Development accounting. American Economic Journal: Macroeconomics, 2(1), 207-23.
Kahle, K. M. (2002). When a buyback isn't a buyback: Open market repurchases and employee options. Journal of Financial Economics, 63(2), 235-261.
Koo, J., & Maeng, K. (2019). Investment opportunities and cash holdings of firms: an empirical study on Korean firms. Applied Economics, 51(23), 2492-2500.
King, R. G., & Levine, R. (1993). Finance and growth: Schumpeter might be right. The quarterly journal of economics, 108(3), 717-737.
Lazonick, W. (2014). Profits without prosperity. Harvard Business Review, 92(9), 46-55.
Lazonick, W. (2015). Buybacks: From Basics to Politics. AIR Special Report, The Academic-Industry Research Network, August, 1.
Markarian, G., & Michenaud, S. (2019). Corporate investment and earnings surprises. The European Journal of Finance, 25(16), 1485-1509.
Nohel, T., & Tarhan, V. (1998). Share repurchases and firm performance:: new evidence on the agency costs of free cash flow. Journal of Financial Economics, 49(2), 187-222.
Sakinc, M. E. (2017). Share repurchases in Europe: a value extraction analysis. The Academic-Industry Research Network Working Paper, 16(2017), 1-27.
Shan, J. Z., Morris, A. G., & Sun, F. (2001). Financial development and economic growth: an egg-and-chicken problem?. Review of international Economics, 9(3), 443-454.
Tobin, J. (1969). A general equilibrium approach to monetary theory. Journal of money, credit and banking, 1(1), 15-29.
Van Nieuwerburgh, S., Buelens, F., & Cuyvers, L. (2006). Stock market development and economic growth in Belgium. Explorations in Economic History, 43(1), 13-38.
Yardeni, E., Abbott, J., Quintana, M., (2020). Corporate Finance Briefing: S&P 500 Buybacks & Dividends. Yardeni Research Inc. https://www.yardeni.com/pub/buybackdiv.pdf
