The effect of planned investment on
investment-cash flow sensitivity of US
airlines during 2005-2010
University of Amsterdam
MSc Business Economics, Finance track
Master Thesis
Author: Shixing Yao
Supervisor: M.A.Dijkstra
Date: July 5, 2015
Abstract
The high investment-cash flow sensitivity caused by agency problem is not a good thing, planned investment might be used to comment this and help to mitigate the problem of attracting external funds during the 2008 financial crisis. This paper uses difference-in-difference analysis to study the effect of planned investment on investment-cash flow sensitivity. The sample of this research is airline industry in the US during period from 2005 to 2010. The analysis reveals that planned investment is positively related with investment-cash flow sensitivity and investment-cash flow sensitivity is more affected by planned investment during the 2008 financial crisis. These results suggest that firms could decrease their planned investment to alleviate the problem of attracting external funds.
Contents
1. Introduction ... 4
2. Literature review ... 5
2.1.
Internal funds and external funds
... 52.2.
Planned investment
... 62.3.
Investment-cash flow sensitivity
... 72.4.
Airlines during the 2008 crisis
... 93. Model and Data ... 10
3.1.
Model
... 103.2.
Data
... 124. Results and analysis ... 14
5. Conclusion ... 20
1. Introduction
Due to the pecking order theory, firms prefer using internal funds to invest. However, agency theory may raise the problem of using internal funds to invest projects of negative NPV and to overinvest, so the sensitivity between investment and cash flow is not always a good thing. Planned investment contains information of project quality, which might be used to measure the access of external funds, so I can use planned investment to comment investment-cash flow sensitivity. This might be especially important during the financial crisis because it is hard to access external funds during that period and this is why I want to research how investment-cash flow sensitivity is affected by planned investment in the US airlines during the 2008 financial crisis.
I choose to research the airline industry because the data of purchase commitment is available. In annual financial reports, airlines would report how much they have committed to purchase in the next few years, which could be used as a proxy of the planned investment. By collecting the amount of committed investments, we can generate a measure of firms’ investment plans, which makes it possible to research the investment-cash flow sensitivity based on planned investment. Moreover, according to Hanlon (2007), travel and tourism is regarded as one of the largest industries in the world and air transport accounts for an important part of this industry. The revenue of airlines is damaged by the 2008 crisis due to the high fuel price and the decline of demands for air transport. (Franke & John, 2011).
I’m going to do difference-in-difference analysis using the sample of airline companies in the US during the period of 2005-2010. First, purchase commitment is used as continuous variable to estimate the effect of planned investment on investment-cash flow sensitivity during the whole period. Then I research the effect of the 2008 crisis on the sensitivity between investment and cash flow. Finally, I use a difference-in-difference-in-difference regression to estimate how investment-cash flow sensitivity is affected by planned investment during the crisis. I also use purchase commitment as dummy variable to do the regressions again to check if the results are the same.
Different from the hypothesis, the results show that planned investment is positively related with investment-cash flow sensitivity, which means the higher planed investment, the less access to external funds. Also, I find that planned investment has more effects on investment-cash flow sensitivity during the 2008 financial crisis. Therefore, in order to alleviate the problem of attracting external funds during the crisis, firms could decrease the amount of their planned investment.
This paper sheds lights on the understanding the implication of planned investment. The main contribution of this work is building the positive relationship between planned investment and investment-cash flow sensitivity, which shows that the higher planned investment, the less access to external funds.
The rest of this paper is organized as follows. Section 2 provides a review of the existing literature, followed by a discussion of the motivation for this paper. Section 3 describes the methodology and the dataset of this study. The results will be given in Section 4 and relate them to earlier hypothesis. Section 5 concludes the paper and offers some directions for further research.
2. Literature review
2.1. Internal funds and external funds
Based on the assumption that in a perfect market, external and internal funds could perfectly substitute for each other, Modigliani and Miller (1958) conclude that the structure of corporate finance would not affect corporate investment decisions. However, perfect capital markets do not exist in the real world, factors like information asymmetry problems, the costs of transaction and agency problem would influence the substitution of external and internal funds, affecting firms’ investment strategies.
Asymmetric information implies that some of the market participants have better or more information than their competitors.Myers and Majluf (1984) state that when a firm wants to raise external funds to invest a project, lenders have less inside information than the manager,
such as the information about the quality of firms’ assets and the quality of projects. Meanwhile, the manager cannot credibly convey the information that the firm’s assets and projects are of good quality, because from the view of lenders, firms with unprofitable projects might also claim that their assets and projects are of good quality in order to attract capital. Therefore, lenders might lend their money to a firm with bad quality or a firm which is going to invest an unprofitable project, so outside investors would require a premium for bearing the risk. In general, the cost of raising external funds increases due to the information asymmetry (Myers, 1984). As a result, financing hierarchy occurs. Companies would prioritize their sources of financing, first preferring internal financing, and then debt, lastly raising equity as a last resort. This is known as pecking order theory.
According to Jensen and Meckling (1976), an agency problem occurs when an agent makes decisions on behalf of the principal and it occurs especially when managers do not own the company. One problem caused by agency problem is that managers may use the internal funds to invest highly risky projects with negative NPV. One of the reasons is that managers’ perquisites increase with investment even they choose to invest project with negative NPV. Another reason for investing negative NPV project is that the value would transfer from the creditor to the shareholder, reducing the total value of the firm but increasing the value of the shareholder. Another problem caused by agency problem is overinvestment, even though there’s no good investment opportunities, managers would still want to use excess cash flow or cash windfalls to make more investments, expanding the size of firm to satisfy their own ambition. Investing projects with negative NPV and overinvesting with internal funds will cause investment more sensitive to cash flow, so high investment-cash flow sensitivity is not always good.
2.2. Planned investment
Purchase commitment is a commitment to purchase a specific number of items from suppliers in the future at a fixed price. A committed purchase can’t be canceled without paying a fee. In a survey organized by the Commerce Department of the US government, during the period from 1947 to 1993, firms are asked to report planned capital expenditure in the next few years.
They find that investment plans account for more than three fourths of the variation in real annual aggregate investment growth.1 Leland et al. (1977) argue that the dedication of internal
funds could be treated as an information of the project quality. They believe that the action of managers who are willing to invest contains information of the project quality, since managers have inside information of firms’ projects. Also, investment plans have substantial forecasting power for returns (Lamont, 2000). Based on a survey conducted by the Commerce Department from 1947 to 1994, he researches the implication of planned investment for stock returns. He argues that planned investment are positively related with current stock returns. Bhana (2008) also believes that planned investment activities may have implications for current and future earnings. He researches 378 cases of investment decision announcements made by South African companies during the period 1995-2004 and finds significant positive excess returns surrounding capital spending announcements, which means markets treat the announcements of investment decisions as a good signal. McConnell et al. (1985) study a couple of firms that announce their planned investment from 1975 to 1981 and find that the announcement of firms’ planned investment contains information of firms’ value. They conclude that an announcement of the increase in planned investment increases the price of firm’s stocks.
A reasonable inference could be made that the more mangers plan to invest in the next few years, the more confidence they have in their projects. Due to the information asymmetry, the amount of committed investment included in firms’ annual reports indicates managers’ willingness to invest and the high planned investment may serve as a signal that the projects are of good quality. Lenders are more willing to lend money to high quality projects, which means that the firm commits to invest a lot in the next few years has more access to external funds.
2.3. Investment-cash flow sensitivity
1 Real actual investment growth is calculated by the formula, where is the nonresidential fixed investment deflator
from the national income accounts.
7
Lamont (1997) focuses on 26 oil companies during the 1986 oil shock. During the sample period, parent firms do not have excess cash flow to subsidize their nonoil segments, these segments need to use their own cash flow to finance investment. The decline of cash flow decreases investment of nonoil segments. Similar results are found by Duchin, Ozbas, and Sensoy (2010), their sample consists of 3668 firms from July 1, 2006 to June 30, 2008. They show that the greater cash holdings a firm has one year before the start of the crisis, the less investment they reduce early in the crisis. They find that the investment of firms with no cash reserves declines 18.5% and the investment of firms with 45.6% of assets in cash does not decline.
Campello et al. (2010) survey 1050 CFOs in the US, Europe and Asia and ask CFOs directly if their firms are constrained by external capital. In the survey, 86% constrained firms’ investment are restricted during the crisis. Constrained firms cut their dividends and prefer to use internal funds to support investment, they might even sell productive assets to raise funds. Furthermore, Chen and Chen (2012) use quarterly data from 2005 to 2009 to test the sensitivity between investment and cash flow during the 2008 financial crisis, they find that investment-cash flow sensitivity disappears during the sample period, but the reasons of the disappearance are still unknown. Since they believe firms have less access to external funds during the crisis, they suggest that investment-cash flow sensitivity is not a good tool to measure firms' access to external funds.
Fazzari et al. (1988) research the investment-cash flow sensitivity based on the ratio of dividends, they use manufacturing 422 firms from year 1969 to 1984, dividing firms into three groups based on their dividend to income ratio. They find that investment-cash flow sensitivity of firms with lower dividends is about two times more than the investment-cash flow sensitivity of high-dividend firms. Kadapakkam et al. (1998) use data from Canada, France, Germany, Great Britain, Japan and the US, during the period from 1982 to 1991. They segment 20428 observations into three groups by firm size. They find that investments of small firms are least sensitive to cash flows. Audretsch & Elston (2002) select their sample within Germany from 1970 to 1986. They divide firms into four different sizes based on the amount of employees and
find that the investment of medium sized firms are most sensitive to their cash flow.Hoshi et al. (1991) focus their study on 145 Japanese manufacturing firms listed on the Tokyo Stock Exchange from 1965 to 1986. Two sets of firms are examined by them, one set is affiliated Keiretsu firms and another one is independent firms. Keiretsu group have a close relationship to the group bank because banks own equities of member firms.2 They find that
Keiretsu-affiliated firms have a low investment-cash flow sensitivity. Francis et al. (2013) study the investment-cash flow sensitivity based on the corporate governance index from a corporate governance rating report.3 They use the average of the summary score in the report to measure
corporate governance and find that firms' investment-cash flow sensitivity increases in response to firms' poor corporate governance.
2.4. Airlines during the 2008 crisis
Franke & John (2011) state that the 2008 crisis damages therevenue of airlines in several ways. First, high fuel prices in 2008 decreased airlines’ profits made in that year,then the downturn of the financial institutions reduced demands for air travel, which affected the business of airlines adversely.
The costs of fuel account for 33% of average operating costs of airlines in 2012, for 22% in 2005 and for 13% in 2001 (IATA, 2012a). Crude oil prices started 2008 at a historic high of $97 a barrel. In February 2008, the speculation in oil drove up oil prices a further 50%, to $147 a barrel. The hike in oil prices naturally caused jet fuel prices to rise, almost 60%, from $114 a barrel at the start of the year to more than $180 a barrel. Fuel costs soared from an average of 28% of airline’s operating costs in 2007 to well over 40% by mid-2008, and to in excess of 50% for some airlines (IATA, 2009 annual report). However, the oil prices decrease significantly by almost 74% at the end of 2008, fall to $36 a barrel. Even though the airlines hedge their fuel
2 A keiretsu group is consist of Japanese firms with close business relationships. Member firms tend to do the
business within the keiretsu group. The financial institution which has a close relationship with keiretsu is the main lender for the group member.
3 The key variables are from a corporate governance rating report, which is published by Credit Lyonnais Securities
Asia (CLSA). The survey contains 495 companies in 25 countries at the end of 2000. Questions in the survey are used to acknowledge information of management discipline, transparency, independence, accountability, responsibility, fairness and social awareness.
9
costs, the volatile fuel prices have an effect on airlines’ profits. According to the annual report of IATA in 2009, fuel hedging contracts signed by airlines indicate that airlines paid much higher prices for their fuel than spot prices during the 2008 fourth quarter.
Moreover, in the 2008 annual report of the IATA, it states that airfreight volumes are always a timely indicator of international trade and economic activity. Both passenger (RPKs) and air freight (FTKs) started to decline during the second quarter of 2008. By December, airfreight volumes had collapsed more than 22% below the level a year earlier. The growth rate of ticket numbers for business and other premium travel had already turned negative by the middle of 2008. Due to the impact of the recession on incomes and confidence, economy travel had fallen more than 5% from year-earlier levels. According to the report of International Air Transport Association (IATA) (2010), the air transportation has declined by more than 6.1% for 2009, it is the largest decrease since World War II.
3. Model and Data
3.1. Model
To explore the effects of planned investment on investment-cash flow, purchase commitment with the interaction terms will be included in the following difference-in-difference regression to estimate:
𝐼𝐼𝐼𝐼𝐼𝐼𝑖𝑖.𝑡𝑡 = 𝛽𝛽0+ 𝛽𝛽1∗ 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽2∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽3∗ 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽4∗ 𝑀𝑀𝑀𝑀𝑖𝑖,𝑡𝑡+ 𝛽𝛽5∗ 𝑙𝑙𝑙𝑙𝐼𝐼𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖,𝑡𝑡
+𝛽𝛽6∗ 𝑠𝑠𝑙𝑙𝑙𝑙𝑙𝑙𝑠𝑠𝑖𝑖,𝑡𝑡+ 𝛼𝛼𝑖𝑖+ 𝜂𝜂𝑡𝑡+ 𝜀𝜀𝑖𝑖,𝑡𝑡 (1)
where 𝑖𝑖 indexes firms, 𝑡𝑡 indexes the year. 𝐼𝐼𝐼𝐼𝐼𝐼𝑖𝑖.𝑡𝑡 is the actual investment scaled by total assets at the beginning of the year. 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡 is the ratio of cash flow gathered by the company to the value of total assets at the beginning of the year. 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡 is the continuous value of purchase commitment scaled by total assets at the beginning of the year, which is used as a proxy of planned investment. Market-to-Book ratio represented by 𝑀𝑀𝑀𝑀𝑖𝑖,𝑡𝑡 is used as a proxy of Tobin’s Q to control the investment opportunity. Other control variables are included in the equation, such as 𝑙𝑙𝑙𝑙𝐼𝐼𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖,𝑡𝑡 and 𝑠𝑠𝑙𝑙𝑙𝑙𝑙𝑙𝑠𝑠𝑖𝑖,𝑡𝑡.Value of all the control variables are scaled by the total assets
at the beginning of the year. 𝛼𝛼𝑖𝑖 is firm fixed effects and 𝜂𝜂𝑡𝑡 is time fixed effects. ε𝑖𝑖,𝑡𝑡 is the error item. The coefficient of interest in equation (1) is the coefficient of the interaction between 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡 with 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡 (𝛽𝛽3). 𝛽𝛽3 represents the effect of planned investment on investment-cash flow
sensitivity during the whole sample period. If it is true that large planned investment could be used as a signal of more access of external funds, then firms with larger planned investment would have lower sensitivity between investment and cash flow, then a negative value of 𝛽𝛽3 is expected.
To estimate the effect of crisis on investment-cash flow sensitivity, the crisis dummy (Crisis) and interaction term (Crisis*Cash flow) will be used in equation (1) instead of the purchase commitment. The value of dummy variable Crisis equals to 1 when the year is larger (not include) than 2007, otherwise it equals to 0. The coefficient of the interaction term measures the difference in investment-cash flow sensitivity between firms during and before the crisis. My hypothesis is that firms would be more rely on their internal funds to make the investment during the 2008 crisis, which means the investment should be more sensitive to the cash flow, so I expected a positive coefficient of the interaction term (Crisis*Cash flow) to be observed.
To estimate the cross effect of planned investment and the 2008 crisis, the following difference-in-difference-in-difference regression is used:
𝐼𝐼𝐼𝐼𝐼𝐼𝑖𝑖.𝑡𝑡 = 𝛽𝛽0+ 𝛽𝛽1∗ CF𝑖𝑖,𝑡𝑡+ 𝛽𝛽2∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽3∗ 𝑐𝑐𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 + 𝛽𝛽4∗ 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽5∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑐𝑐𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠
+𝛽𝛽6∗ 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑐𝑐𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 + 𝛽𝛽7∗ 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑐𝑐𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 ∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡+ 𝛽𝛽8∗ 𝑀𝑀𝑀𝑀𝑖𝑖,𝑡𝑡+ 𝛽𝛽9∗ 𝑙𝑙𝑙𝑙𝐼𝐼𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖,𝑡𝑡
+𝛽𝛽10∗ 𝑠𝑠𝑙𝑙𝑙𝑙𝑙𝑙𝑠𝑠𝑖𝑖,𝑡𝑡+ 𝛼𝛼𝑖𝑖+ 𝜂𝜂𝑡𝑡+ 𝜀𝜀𝑖𝑖,𝑡𝑡 (2)
The coefficient of interaction term 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑐𝑐𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 ∗ 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡 (𝛽𝛽7) measures the difference between the effect of planned investment on investment-cash flow sensitivity during and before the crisis. So if planned investment has more effects on investment-cash flow sensitivity during the crisis than it has before the crisis, the value of 𝛽𝛽7 would be positive.
Finally, equation (1) and (2) mentioned above will be estimated again with the dummy variable of purchase commitment (𝑃𝑃𝐶𝐶) instead of continuous value (𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡). If a firm’s value of purchase
commitment scaled by total assets at the beginning is larger than the median value of the whole sample, the firm belongs to the group that firms with large planned investment and the dummy variable 𝑃𝑃𝐶𝐶 equals to 1. The coefficient of interaction term ( 𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑃𝑃𝐶𝐶) measures the difference of investment-cash flow sensitivity between firms with large planned investment and firms with small planned investment during the period 2005-2010. If it is true that large planned investment could be used as a signal of more access of external funds, then firms with larger planned investment would have lower sensitivity between investment and cash flow, the coefficient of interaction term (𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡∗ 𝑃𝑃𝐶𝐶) is expected to be negative. Similar to equation (2), (𝛽𝛽6+ 𝛽𝛽7) measures the difference between investment-cash flow sensitivity for firms with large planned investment during and before crisis, which is expected to be positive. 𝛽𝛽6 measures the difference between investment-cash flow sensitivity for firms with small planned investment during and before crisis, which is expected to be positive. 𝛽𝛽4 measures the difference of investment-cash flow sensitivity between firms with large planned investment and firms with small planned investment before crisis, which is expected to be negative. (𝛽𝛽4+ 𝛽𝛽7) measures the difference of investment-cash flow sensitivity between firms with large investment and firms with small planned investment during the crisis, which is expected to be negative. 𝛽𝛽7 is expected to be positive which means that difference in investment-cash flow sensitivity between firm with large planned investment and firms with small planned investment during the crisis is larger than the difference before the crisis.
3.2. Data
The data of the US airline companies through 2005 to 2010 are gathered from the website of U.S. Securities and Exchange Commission (SEC)4. 174 airlines are included within the code
of 4512 and 4522. Publicly-traded airline companies typically report the amount of committed investment for the next few years directly in their 10-K filings. The airlines’ 10-K filings could be found from SEC-filings, the official websites of companies or Thomson Reuters Eikon. The data of committed investments for each year from 2005-2010 were hand-collected from the
4 Firms are collected if they belong to the SIC code of 4512 (AIR TRANSPORTATION, SCHEDULED) or 4522
(AIR TRANSPORTATION, NONSCHEDULED). 12
annual report (10-K) of airlines in the US for the fiscal year from 2004 to 2009. The data could be found in the footnote to financial statement called “commitments and contingencies” or some footnotes with similar titles, it is basically the amount of money used to purchase aircrafts (purchase commitment). It is composed of two parts which are aircraft purchase commitments and other purchase commitments. Airlines without reporting their purchase commitment are excluded from the sample. After collecting the purchase commitments (in million USD), 28 airlines are contained in the sample. The main variables and control variables could be found in the standard database COMPUSTAT, over the period 2004-2010. Actual investment is measured by capital expenditure. The contemporaneous cash flow is the sum of the income before extraordinary item and the depreciation. The Market-to-Book ratio of a firm is the ratio of a firm’s market value over its book value, which is using market value of equity plus book value of assets minus book value of equity minus deferred taxes, divided by book value of assets. Leverage is calculated as the ratio of total liabilities to total assets at the beginning of the year. Net sales divided by total assets at the beginning of the year is computed as sales.
Table 1. Summary statistics: Airlines with small planned investment, large planned investment and the total airlines
This table describes the value of regression variables. Total assets in this table represents the total assets at the beginning of the year. All the variables are not scaled by total assets at the beginning of the year.
Airlines with small planned investment Airlines with large planned investment Total airlines
Variable mean mean mean min max sd
Investment 356.2804 365.7134 360.9969 1.6550 1608 400.2009 Cash flow 108.2973 120.4964 114.3968 -20165 23546 4287.5919 MB 1.2534 1.1657 1.2063 0.5926 2.3050 0.3722 Leverage 0.9411 0.8212 0.8811 0.2734 2.4174 0.3985 Sales 10354.5281 3901.2936 7127.9109 115.6190 28063 8208.6228 PC 211.0994 442.5612 321.5215 0.0000 1520 372.7884 Total assets 12617.3897 4379.5199 8498.4548 49.0700 43188 10201.2688 N 57 57 114 13
Observations with leverage larger than 4 are eliminated and all the variables are winsorized. This selection procedure finally leaves me 114 observations of airlines during period 2005-2010. After the selection procedure above, the sample period from 2005 to 2010 but with gaps, so the data I acquire are unbalanced panel data.
Tables 1 presents the description for the full sample data. The first two columns in table one revel the mean value of variables for two groups which are separated based on the value of purchase commitment scaled by total assets at the beginning of the year. The actual investment of airlines with large planned investment is slightly higher than that of airlines with small planned investment. However, the total assets at the beginning of the year for airlines with small planned investment is almost three times as large as that for airlines with large planned investment. So if take total assets into account, the difference in actual investment between these two groups is obvious. Leverage for airlines with small planned investment is slightly larger than the leverage for another group, which means based on total assets, airlines with small planned investment use more liabilities. Total assets and sales could be used to indicate the size of firms, so the value of these two variables indicate airlines with small planned investment have larger size. According to the conclusion of Kadapakkam et al. (1998), investments of small firms are least sensitive to cash flows, the inference for table 1 is airlines with large planned investment has small investment cash flow sensitivity. Since average value of variables cannot reflect all the information of variables, the inference still need to be tested.
4. Results and analysis
This section examines how planned investment and the 2008 crisis affect the investment-cash flow sensitivity. First, continuous variable 𝑃𝑃𝐶𝐶𝑖𝑖,𝑡𝑡 will be used to do the different-in-different and difference-in-difference-in-difference regressions. Then the dummy variable 𝑃𝑃𝐶𝐶 will be used to do the regressions again. The results and analysis are presented in table 2, table 3 and table 4.
Table 2. The effect of planned investment and the effect of Crisis on investment-cash flow sensitivity
This table reports the results of difference-in-difference regressions.Column 1, 2 and 3 estimate the effect of planned investment on investment-cash flow sensitivity. Column 4, 5 and 6 report the difference of investment-cash flow sensitivity during and before the crisis. The independent variable is 𝐼𝐼𝐼𝐼𝐼𝐼, calculated as the capital expenditure scaled by total assets at the beginning of the year. Crisis is dummy variable and equals to 1 if. All variables are defined in Model and Data. Firm fixed effects and year fixed effects are included to control omitted variables.The sample period is 2005 to 2010. Constants were included in the regressions but are not reported. All standard errors are clustered at the firm level and t-statistics are reported in parentheses. *, **, and*** indicate significance at 5%, 1% and 0.1%, respectively.
Effects of Planned Investment Effects of Crisis
(1) (2) (3) (4) (5) (6) Cash flow 0.0068 0.0301 -0.0449 0.0049 -0.0049 -0.0367 (0.46) (0.78) (-1.46) (0.21) (-0.13) (-1.10) PC 0.358** * 0.367*** 0.344* (3.98) (3.66) (2.34) Cash flow*PC 2.000* 1.900** 2.129** (2.60) (2.66) (3.30) Crisis -0.0165 -0.0337 -0.119** (-1.04) (-1.50) (-3.69) Cash flow*Crisis 0.419* 0.493* 0.516* (2.06) (2.34) (2.61) Market-to-book ratio 0.0222 -0.0798 -0.0323 -0.0915 (0.92) (-1.84) (-1.30) (-1.75) Leverage 0.0205 0.0321 0.0087 0.0461 (1.31) (1.72) (0.44) (1.53) Sales 0.0054 0.0274 0.0146 0.0330 (0.35) (1.49) (1.13) (1.35)
Firm fixed effects No No Yes No No Yes
Time fixed effects No No Yes No No Yes
N 109 89 89 114 93 93
0.4374 0.4603 0.5841 0.1865 0.2227 0.3913
Table 2 presents the results of the difference-in-difference regressions, estimating the effects of planned investment and crisis on investment-cash flow sensitivity respectively. Column 1 to 3 reveal that the coefficients of the interaction term are all positive and significantly different form 0, which means the larger planned investment, the higher investment-cash flow sensitivity. The results are different from my expectation. The positive relationship between planned investment and investment-cash flow sensitivity indicates that firms with larger planned investment have less access to the external funds. Column 4 to 6 estimate the effect of the 2008 crisis on sensitivity between investment and cash flow. Coefficients of the interaction term range from 0.419to 0.516, which is positive and significant. The positive coefficients reveal that investment is more sensitive to cash flow during the crisis than that before the crisis. Therefore, during the crisis firms have less access to external funds. This result is consistent with the founding in Campello et al. (2010) that firms are financially constrained during the crisis and need to use internal capital to finance investment.
According to the column 3, 4 and 5 in table 3, the coefficients of interaction term 𝐶𝐶𝑙𝑙𝑠𝑠ℎ 𝑓𝑓𝑙𝑙𝑓𝑓𝑓𝑓 ∗ 𝑃𝑃𝐶𝐶 ∗ 𝐶𝐶𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 are positive, which is the same as expected in the model part. It represents that the effect of planned investment on investment-cash flow sensitivity during the crisis is larger than the effect of planned investment on investment-cash flow sensitivity before the crisis. The positive coefficients of 𝐶𝐶𝑙𝑙𝑠𝑠ℎ 𝑓𝑓𝑙𝑙𝑓𝑓𝑓𝑓 ∗ 𝑃𝑃𝐶𝐶 and 𝐶𝐶𝑙𝑙𝑠𝑠ℎ 𝑓𝑓𝑙𝑙𝑓𝑓𝑓𝑓 ∗ 𝐶𝐶𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 in column 3, 4 and 5 reveal the same results in column 1 and 2, which are estimated by using difference-in-difference regression. In column 5, the effect of planned investment on investment-cash flow sensitivity during and before the crisis is 1.847, which is calculated as (1.24+0.607), and 0.607 respectively. Therefore, an increase on planned investment during the crisis will make a firm has less access to external funds and the decline of the external funds is larger than that when a firm increase the same amount planned investment before the crisis. During the crisis, a firm could decrease their planned investment to alleviate the problem of attracting external funds.
Table 3. The cross effect of planned investment and crisis on investment-cash flow sensitivity
This table reports the results of difference-in-difference-in-difference regressions. Column 1 and 2 is the column 3and 6 in table 2. Column 3, 4 and 5 estimate difference between the effect of planned investment on investment-cash flow sensitivity during and before the crisis. The independent variable is, calculated as the capital expenditure scaled by total assets at the beginning of the year. Crisis is dummy variable and equals to 1 if. All variables are defined in Model and Data. Firm fixed effects and year fixed effects are included to control omitted variables. The sample period is 2005 to 2010. Constants were included in the regressions but are not reported. All standard errors are clustered at the firm level and t-statistics are reported in parentheses. *, **, and*** indicate significance at 5%, 1% and 0.1%, respectively.
Diff-in-Diff Diff-in-Diff-in-Diff (1) (2) (3) (4) (5) Cash flow -0.0449 -0.0367 0.0011 0.0164 -0.0242 (-1.46) (-1.10) (0.09) (0.55) (-0.63) PC 0.344* 0.412** 0.487** 0.489* (2.34) (3.18) (2.95) (2.44) Crisis -0.119** -0.0141 -0.0347 -0.0667* (-3.69) (-1.06) (-1.79) (-2.56) Cash flow *PC 2.129** 0.423 0.392 0.607 (3.30) (0.46) (0.38) (0.60) Cash flow*Crisis 0.516* 0.196 0.145 0.207 (2.61) (1.73) (1.08) (1.24) PC*Crisis -0.0383 -0.0318 -0.155 (-0.21) (-0.16) (-0.75) Cash flow*PC*Crisis 1.511 1.618 1.248 (1.18) (1.31) (1.08)
Control variables5 Yes Yes No Yes Yes
Firm fixed effects Yes Yes No No Yes
Time fixed effects Yes Yes No No Yes
N 89 93 109 89 89
0.5841 0.3913 0.5121 0.5935 0.6312
5 Control variables include Market-to-Book ratio, leverage and sales.
17
Table 4. The investment-cash flow sensitivity of firms with large planned investment during the 2008 crisis
This table reports the results of difference-in-difference and difference-in-difference-in-difference regressions. PC is a dummy variable and equals to 1 if the firm has large planned investment. The independent variable is 𝐼𝐼𝐼𝐼𝐼𝐼, calculated as the capital expenditure scaled by total assets at the beginning of the year. 𝐶𝐶𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 is dummy variable and equals to 1 if. All variables are defined in Model and Data. Firm fixed effects and year fixed effects are included to control omitted variables. The sample period is 2005 to 2010. Constants were included in the regressions but are not reported. All standard errors are clustered at the firm level and t-statistics are reported in parentheses. *, **, and*** indicate significance at 5%, 1% and 0.1%, respectively. Diff-in-Diff Diff-in-Diff-in-Diff (1) (2) (3) (4) (5) (6) Cash flow 0.0079 0.0072 -0.0628* 0.0017 -0.0050 -0.0504 (0.70) (0.29) (-2.35) (0.16) (-0.21) (-1.76) PC 0.0474*** 0.0456** 0.0330** 0.0540*** 0.0491** 0.0505* (3.78) (3.32) (3.17) (3.48) (2.82) (2.50) Cash flow*PC 0.418* 0.418** 0.504*** 0.122 0.176 0.156 (2.29) (2.71) (4.29) (1.11) (1.30) (1.18) Crisis -0.0043 -0.0158 -0.0831** (-0.51) (-1.07) (-2.99) Cash flow*Crisis 0.0953* 0.1330 0.0554 (1.99) (1.39) (0.41) PC*Crisis -0.0137 -0.0081 -0.0539 (-0.48) (-0.24) (-1.72) Cash flow*PC*Crisis 0.390 0.312 0.501** (1.64) (1.25) (3.01)
Control variables6 No Yes Yes No Yes Yes
Firm fixed effects No No Yes No No Yes
Time fixed effects No No Yes No No Yes
N 114 93 93 114 93 93
0.3477 0.3667 0.5179 0.4050 0.4121 0.5791
6 Control variables include Market-to-Book ratio, leverage and sales.
18
Different from table 2 and table 3, instead of using continuous variable 𝑃𝑃𝐶𝐶, dummy variable 𝑃𝑃𝐶𝐶 is used in table 4. Column 1, 2 and 3 show the difference in investment-cash flow sensitivity between firms with large planned investment and firms with small planned investment. The coefficients of interaction term 𝐶𝐶𝑙𝑙𝑠𝑠ℎ 𝑓𝑓𝑙𝑙𝑓𝑓𝑓𝑓 ∗ 𝑃𝑃𝐶𝐶 are all positive and significantly different form 0 in the first three columns, which indicates that compared with firms with small planned investment, firms with large planned investment have higher investment-cash flow sensitivity. This result is consistent with the result of table 2, which uses continuous variable of purchase commitment.
Dummy variable 𝐶𝐶𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 is included in column 4, 5 and 6 to estimate the change of investment-cash flow sensitivity between firms with large and small planned investment before and during the crisis. 0.0554 and (0.501+0.0554) in column 6 indicate that no matter how much firms plan to invest, investment is more sensitive to cash flow during the crisis. 0.156 and (0.156+0.501) indicate that no matter when, before or during the crisis, investment is more sensitive to cash flow for firms with large planned investment. In column 6, the positive and significant coefficient of interaction term 𝐶𝐶𝑙𝑙𝑠𝑠ℎ 𝑓𝑓𝑙𝑙𝑓𝑓𝑓𝑓 ∗ 𝑃𝑃𝐶𝐶 ∗ 𝐶𝐶𝑙𝑙𝑖𝑖𝑠𝑠𝑖𝑖𝑠𝑠 means that difference in investment-cash flow sensitivity between firms with large planned investment and firms with small planned investment during the crisis is larger than the difference before the crisis. This could be used to indicate that the planned investment has more effects on investment-cash flow sensitivity during the crisis. Results in table 4 is consistent with results in table 3, which uses continuous variable of purchase commitment to do the diff-in-diff-in-diff regressions.
To summarize the results in all the three tables, high planned investment cannot be interpreted as more access to external funds due to the positive relationship between planned investment and investment-cash flow sensitivity, which means the larger planned investment, the less access to external funds. However, it is true that during the crisis, firms have less access to external funds and planned investment has more effects on investment-cash flow sensitivity during the crisis. Therefore, firms could decrease their planned investment to alleviate the problem of attracting external funds during the crisis.
5. Conclusion
This paper uses difference-in-difference analysis to study the sample of airlines in the US during period 2005-2010. Purchase commitment in firms’ annual reports offers me the opportunity to investigate the effect of firms’ planned investment on the sensitivity between investment and cash flow before and during the 2008 crisis. In this paper, purchase commitment is used as continuous variable and dummy variable respectively in the regressions and the same result is found by using these two variables.
McConnell et al. (1985) and Bhana (2008) find a positive relationship between the planned investment and stock return, so they believe that the markets treat the announcement of planned investment as a good news. Different form their conclusions, in this paper, planned investment was found to have a positive relationship with investment-cash flow sensitivity during the whole period, which means that firms with larger planned investment have less access to external funds. Therefore, outside investors prefer to lend their money to firms with small planned investment compared with firms have large planned investment.
Secondly, the investment of firms is more sensitive to cash flow during the crisis, which means firms have less access to external funds during the crisis. This result is consistent with the findings of Campello et al. (2010), which states that during the crisis firms’ investment are restricted by financial constraints.
Finally, the planned investment has more effects on investment-cash flow sensitivity during the crisis. So it is more effective for firms to decrease their planned investment to mitigate the problem of attracting external funds during the crisis.
Appendix:
Table 5. Correlation analysis
This table reports correlations of variables.
variable Investment Cash flow MB leverage sales PC Investment 1.0000 Cash flow 0.2093 1.0000 MB 0.0723 -0.0389 1.0000 leverage 0.0192 -0.2140 0.0918 1.0000 sales 0.2118 0.1140 0.1531 0.0648 1.0000 PC 0.4412 -0.0885 -0.2464 -0.0604 0.3672 1.0000
Reference
Audretsch, D. B., & Elston, J. A. (2002). Does firm size matter? Evidence on the impact of liquidity constraints on firm investment behavior in Germany. International Journal of Industrial Organization, 20(1), 1-17.
Bhana, N. (2008). The market reaction to capital expenditure announcements. Investment Analysts Journal, (68), 57-68.
Brealey, R., Leland, H. E., & Pyle, D. H. (1977). Informational asymmetries, financial structure, and financial intermediation. The journal of Finance, 32(2), 371-387.
Campello, M., Graham, J. R., & Harvey, C. R. (2010). The real effects of financial constraints: Evidence from a financial crisis. Journal of Financial Economics, 97(3), 470-487.
Chen, H. J., & Chen, S. J. (2012). Investment-cash flow sensitivity cannot be a good measure of financial constraints: Evidence from the time series. Journal of Financial Economics, 103(2), 393-410.
Duchin, R., Ozbas, O., & Sensoy, B. A. (2010). Costly external finance, corporate investment, and the subprime mortgage credit crisis. Journal of Financial Economics, 97(3), 418-435. Fazzari, S. M., Hubbard, R. G., & Petersen, B. С. (1988). Financing Constraints and Corporate Investment. Brookings Papers on Economic Activity, 1988(1), 141-195.
Francis, B., Hasan, I., Song, L., & Waisman, M. (2013). Corporate governance and investment-cash flow sensitivity: Evidence from emerging markets. Emerging Markets Review, 15, 57-71.
Franke, M., & John, F. (2011). What comes next after recession?–Airline industry scenarios and potential end games. Journal of Air Transport Management, 17(1), 19-26.
Hoshi, T., Kashyap, A., & Scharfstein, D. (1991). Corporate structure, liquidity, and investment: Evidence from Japanese industrial groups. The Quarterly Journal of Economics, 33-60.
IATA, annual review. [Online] Available from: http://www.iata.org/publications/Pages/annual-review.aspx
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Kadapakkam, P. R., Kumar, P. C., & Riddick, L. A. (1998). The impact of cash flows and firm size on investment: The international evidence. Journal of banking & Finance, 22(3), 293-320.
Lamont, O. (1997). Cash flow and investment: Evidence from internal capital markets. The Journal of Finance, 52(1), 83-109.
Lamont, O. A. (2000). Investment plans and stock returns. The Journal of Finance, 55(6), 2719-2745.
McConnell, J. J., & Muscarella, C. J. (1985). Corporate capital expenditure decisions and the market value of the firm. Journal of financial economics, 14(3), 399-422.
Modigliani, F., & Miller, M. H. (1958). The cost of capital, corporation finance and the theory of investment. The American economic review, 261-297.
Myers, S. C. (1984). The capital structure puzzle. The journal of finance, 39(3), 574-592.
Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of financial economics, 13(2), 187-221.
