I’m attending now the 27th annual Cato Institute Monetary Conference.
Just before the conference kicked off this morning at 9am (EST), The Economist‘s Zanny Minton Beddoes suggested that I do some live blogging while here. I hadn’t thought to do that. (I’m here to moderate one of the afternoon panels.)
But, what the heck, here’s at least one post coming live at you from the conference……
….
Following an outstanding opening address by Allan Meltzer, the first panel featured Zanny Minton Beddoes, George Melloan, Benn Steil, and Peter Wallison. Fine presentations all.
In his remarks, Benn Steil mentioned a fact that I was (to my chagrin) unaware of, but it strikes me – as it strikes Steil – as being one important reason for excessive build-up of debt in America. According to page 8 of this 2005 CBO study,
The resulting [tax] rate on equity-financed corporate capital income is 36.1 percent and that on debt-financed corporate capital income is -6.4 percent, a difference of 42.5 percentage points. The rate on equity-financed corporate capital income is higher than the statutory corporate tax rate because of the extra tax imposed on dividends and capital gains at the individual level.
Such a huge deviation from neutrality in taxing equity-financed corporate income and in taxing debt-financed corporate income cannot help but to distort financing decisions toward debt – perhaps dangerously so.
Podcast RSS Feed
Full EconTalk Text
{ 31 comments }
But what is the actual cost of money for debt financed corporate capital? Is suspect it is still absolutely lower, but not as drastically. Figuring in opportunity costs when tying up cash in the bank is what makes this a no brainer for companies like Microsoft to make their first forays into the debt markets (September, 2008). What’s really interesting about debt offerings, which are heavily regulated by the SEC of course, is that even though the companies have to state a purpose for the funds, the money raised frees up other money for other purposes. So there was speculation Microsoft was raising capital for general operations in order to be able to use its cash on hand to acquire SAP. Corporate sleight of hand aside, the downstream effects of encouraging very profitable companies to play in the debt markets don’t seem good.
Out of curiosity, do you know how long this disparity has persisted? I’ve heard this explanation several times before too and I certainly share your concerns.I’m always uncomfortable about assigning an explanation of the run up of privately held debt in the last decade or so to something that has been a part of the scene for several decades. The distortion comes from differences between depreciation rules in tax law and actual depreciation – if that difference widened considerably prior to the bubble, that would be powerful evidence.
Not sure that accelerated depreciation would favor debt financing over equity financing. Perhaps I’m too tired from noontime basketball to think through this carefully. Can you explain?
As for the historical use of accelerated depreciation, Accelerated Cost Recovery System (ACRS) was implemented in 1975 and Modified Accelerated Cost Recovery System (MACRS) was implemeted in 1986. MACRS allowed even faster depreciation of capital assets.
Certainly the interest tax shield advantage of debt has existed for decades – maybe since 1894 when the U.S. government instituted a separate tax system for corporations. I suspect it wasn’t used as much as it should have between 1940 and 1980 because corporate decision-makers had lived through the Great Depression.
Not to be a hot dog but I think ACRS came in 1981.
Not to be a hot dog but I think ACRS came in 1981.
Thanks for the correction. I misread my source, and forgot that it was part of Kemp-Roth, signed during Reagan’s first year.
Thanks for the correction. I misread my source, and forgot that it was part of Kemp-Roth, signed during Reagan’s first year.
On the surface, this shows how manipulation of the tax code creates problems. I wonder though if corporations played a role, via lobbying, for just such an outcome. Regardless, elimination of corporate taxation would place a check on both inept politicians and manipulative corporations.
For 50 years we’ve heard debates about Modigliani-Miller, which says that degree of leverage does not change the value of the firm. But Modigliani-Miller, which was always only a theoretical conclusion, carries three assumptions which make its applicability questionable:
1. no taxes
2. no bankruptcy costs
3. efficient markets
The last may be debatable, but the first two are not. The question has thus been at what point does bankruptcy risk overcome the value of interest tax shields.
Michael Milkin and others correctly concluded in the 1970’s and 1980’s that most U.S. firms were underleveraged – at least if the goal of corporate leaders was to maximize shareholder value. At the same time, they realized that leveraged buyouts were efficient tools for removing underperforming management. The retirement of equity and the expansion of firms through issuance of debt exploded during the 1980’s and 1990’s as a result of their thinking.
As I see it, leveraging of corporations is highly desirable given the tax advantage. But it carries with it the executive concern over interest coverage, an issue which wasn’t so important 40 and 50 years ago. Some have argued that such concern has led to a short term focus in decision-making.
The interest tax shield is not the only government intervention which distorts capital structure decisions.
Sarbanes-Oxley provided additional impetus for debt financing. Corporate leaders recognized that by taking their firms private, they could escape the regulatory burdens and the criminal liabilities implemented with SOX. The leveraged buyout was the tool used to retire enough equity so that the form could be taken private.
The interest tax shield is not the only government intervention which distorts capital structure decisions.
Sarbanes-Oxley provided additional impetus for debt financing. Corporate leaders recognized that by taking their firms private, they could escape the regulatory burdens and the criminal liabilities implemented with SOX. The leveraged buyout was the tool used to retire enough equity so that the form could be taken private.
What other econ/finance/gov blogs do you read?
I regularly read Marginal Revolution (Econ Professors Tyler Cowen and Alex Tabarrok) and Digital Rules (Forbes Publisher Rich karlgaard). I occasionally read Coyote Blog (entrepreneur Warren Meyer) and Greg Mankiw’s blog.
I regularly read Marginal Revolution (Econ Professors Tyler Cowen and Alex Tabarrok) and Digital Rules (Forbes Publisher Rich karlgaard). I occasionally read Coyote Blog (entrepreneur Warren Meyer) and Greg Mankiw’s blog.
What other econ/finance/gov blogs do you read?
Economists Steven Kaplan and Per Strömberg suggest that increasing leverage can sometimes provide benefits in addition to interest tax shields:
“Leverage creates pressure on managers not to waste money, because they must make interest and principal payments. This pressure reduces the “free cash flow” problems described in Jensen (1986), in which management teams in mature industries with weak corporate governance had many ways in which they could dissipate these funds rather than returning them to investors.”
Economists Steven Kaplan and Per Strömberg suggest that increasing leverage can sometimes provide benefits in addition to interest tax shields:
“Leverage creates pressure on managers not to waste money, because they must make interest and principal payments. This pressure reduces the “free cash flow” problems described in Jensen (1986), in which management teams in mature industries with weak corporate governance had many ways in which they could dissipate these funds rather than returning them to investors.”
Long-run debt ratios (long-term debt over book total assets) amongst public companies in the U.S. is very stable from 1978 (the earliest year I checked) through 2007 (the latest year). There have been several tax regime shifts during this period that affect the relative value of debt financing from a tax perspective. Depreciation regulations have been stable.
The trade-off theory of corporate finance suggests that firms choose their debt ratios by balancing the tax benefits of debt financing with the potential costs of bankruptcy. The presence of large depreciation charges reduces the tax benefits of debt, as do investment tax credits.
The pecking order of finance posits that firms prefer internal financing, then debt financing, and then equity financing last, due to the negative signals equity financing sends.
While most tests of leverage choices by firms are difficult and open to interpretation, the pecking order theory tends to hold up better. This is consistent with the debt being a governance substitute, as John Dewey pointed out.
In short, while this disparity in financing costs is large, it appears to be of secondary importance to most corporations and so not as big of a concern here as the effect of corporate tax rates on capital formation in the first place.
Interesting comments, neoaustrian. I especially appreciate the paragraphs about trade-off theory and pecking order.I’m surprised to read that LT debt to book total assets has been stable. Could it be that book values were not valid during the higher inflation period of the late 1970′s and early 1980′s?I just found a 1999 article from the New York Fed which suggests that corporate leverage declined during the 1990′s. Their alternate measure of leverage – total debt to total assets, weighted by market value – does show that leverage was relatively stable from 1974 through 1999. They do acknowledge, though, that:“When prices are rising, the value of assets tends to be too lowto provide a good gauge of future earnings capacitybecause that value includes assets ourchased years ago at prices below current prices.” That article also shows total debt to firm market value, weighted by stock market value. This measure seems to reveal that leverage declined sharply i the last half of the 1990′s. But if equities were overvalued in the late 1990′s, then leverage would have been understated.
You highlight a couple of good issues, regarding the data, that are important. First, book values tend to become outdated. This is true, but I’ll come back to why it isn’t a big concern here.
Second is the use of market value. Capturing a balance sheet item at any given time using market value can be distortionary because the market value at any given time could be significantly different from the true (unknowable) value of the firm. So, if one is using market values, one must use a long-run average to try and balance out distortions. I didn’t read the Fed paper, so I’m not sure if or how they address the issue.
I see the Fed paper used total debt. The concern in this article is about the tax differential, so total debt is not the correct metric. The reason is that total debt includes short-term debt (like notes payable) and trade credit (accounts payable). Interest on these items is not tax deductible, and/or these items are tactical rather than strategic. Long-term debt interest is tax deductible, and long-term debt is a strategic item, so it is considered to be a corporate strategy decision of executive management.
Now, to return to the issue of book assets: inflation is definitely a concern here, because replacement values of assets will, on average, be higher than historical costs. This serves to bias debt ratios downwards, however, rather than upwards, because the denominator is artificially low. The other thing is that net capital expenditure (capital expenditure – depreciation) is positive for the most part during this period. This indicates that most firms are replacing assets as they wear (sp?) out, so that total assets are not likely to be extremely far from replacement value.
neoaustrian: “. The reason is that total debt includes short-term debt (like notes payable) and trade credit (accounts payable). Interest on these items is not tax deductible”
I don’t thinik that is correct, neoaustrian. Are you referring to U.S. federal taxation of corporations? Here’s a primer on tax treatment of business interest expense.
neoaustrian: “This serves to bias debt ratios downwards, however, rather than upwards, because the denominator is artificially low.”
I agree that inflation makes the denominator artificailly low. That was the point made by me and by the economists of the New York Fed. If the denominator is artificially low, the debt ratio is thus artificially high. That was the case in the high inflation period of the 1970′s and early 1980′s.
neoaustrian: “. The reason is that total debt includes short-term debt (like notes payable) and trade credit (accounts payable). Interest on these items is not tax deductible”
I don’t thinik that is correct, neoaustrian. Are you referring to U.S. federal taxation of corporations? Here’s a primer on tax treatment of business interest expense.
neoaustrian: “This serves to bias debt ratios downwards, however, rather than upwards, because the denominator is artificially low.”
I agree that inflation makes the denominator artificailly low. That was the point made by me and by the economists of the New York Fed. If the denominator is artificially low, the debt ratio is thus artificially high. That was the case in the high inflation period of the 1970′s and early 1980′s.
Sorry, I was unclear in the first comment you quote. Short-term debt interest is a tax deduction. Trade credit largely isn’t, because very few corporations pay interest. Also included (I forgot to make this clear) in total debt are accruals, which include wages earned but unpaid. No interest on that either.
On the second point: I’m not sure where my head was at. It is obvious that lower denominators result in higher ratios. Thank you for pointing that out.
When I get a minute, I’ll deflate the total assets and long-term debt by the PPI (or would CPI be better?) and see what that tells us. Incidentally, did the Fed paper deflate market values by CPI or PPI?
Sorry, I was unclear in the first comment you quote. Short-term debt interest is a tax deduction. Trade credit largely isn’t, because very few corporations pay interest. Also included (I forgot to make this clear) in total debt are accruals, which include wages earned but unpaid. No interest on that either.
On the second point: I’m not sure where my head was at. It is obvious that lower denominators result in higher ratios. Thank you for pointing that out.
When I get a minute, I’ll deflate the total assets and long-term debt by the PPI (or would CPI be better?) and see what that tells us. Incidentally, did the Fed paper deflate market values by CPI or PPI?
You highlight a couple of good issues, regarding the data, that are important. First, book values tend to become outdated. This is true, but I’ll come back to why it isn’t a big concern here.
Second is the use of market value. Capturing a balance sheet item at any given time using market value can be distortionary because the market value at any given time could be significantly different from the true (unknowable) value of the firm. So, if one is using market values, one must use a long-run average to try and balance out distortions. I didn’t read the Fed paper, so I’m not sure if or how they address the issue.
I see the Fed paper used total debt. The concern in this article is about the tax differential, so total debt is not the correct metric. The reason is that total debt includes short-term debt (like notes payable) and trade credit (accounts payable). Interest on these items is not tax deductible, and/or these items are tactical rather than strategic. Long-term debt interest is tax deductible, and long-term debt is a strategic item, so it is considered to be a corporate strategy decision of executive management.
Now, to return to the issue of book assets: inflation is definitely a concern here, because replacement values of assets will, on average, be higher than historical costs. This serves to bias debt ratios downwards, however, rather than upwards, because the denominator is artificially low. The other thing is that net capital expenditure (capital expenditure – depreciation) is positive for the most part during this period. This indicates that most firms are replacing assets as they wear (sp?) out, so that total assets are not likely to be extremely far from replacement value.
Interesting comments, neoaustrian. I especially appreciate the paragraphs about trade-off theory and pecking order.I’m surprised to read that LT debt to book total assets has been stable. Could it be that book values were not valid during the higher inflation period of the late 1970′s and early 1980′s?I just found a 1999 article from the New York Fed which suggests that corporate leverage declined during the 1990′s. Their alternate measure of leverage – total debt to total assets, weighted by market value – does show that leverage was relatively stable from 1974 through 1999. They do acknowledge, though, that:“When prices are rising, the value of assets tends to be too lowto provide a good gauge of future earnings capacitybecause that value includes assets ourchased years ago at prices below current prices.” That article also shows total debt to firm market value, weighted by stock market value. This measure seems to reveal that leverage declined sharply i the last half of the 1990′s. But if equities were overvalued in the late 1990′s, then leverage would have been understated.
Long-run debt ratios (long-term debt over book total assets) amongst public companies in the U.S. is very stable from 1978 (the earliest year I checked) through 2007 (the latest year). There have been several tax regime shifts during this period that affect the relative value of debt financing from a tax perspective. Depreciation regulations have been stable.
The trade-off theory of corporate finance suggests that firms choose their debt ratios by balancing the tax benefits of debt financing with the potential costs of bankruptcy. The presence of large depreciation charges reduces the tax benefits of debt, as do investment tax credits.
The pecking order of finance posits that firms prefer internal financing, then debt financing, and then equity financing last, due to the negative signals equity financing sends.
While most tests of leverage choices by firms are difficult and open to interpretation, the pecking order theory tends to hold up better. This is consistent with the debt being a governance substitute, as John Dewey pointed out.
In short, while this disparity in financing costs is large, it appears to be of secondary importance to most corporations and so not as big of a concern here as the effect of corporate tax rates on capital formation in the first place.
“I’ll deflate the total assets and long-term debt by the PPI”
I’m not sure what that will tell you. As I see it, the distortion for debt ratios is caused by assets and debt on the books being acquired at different periods. If the book values of assets represent sums paid for those assets many years earlier – and the debt refers to debt acquired relatively recently – then the ratio of debt to assets is distorted. That has been exactly the case at large companies at which I’ve worked. Debt was extended to us not on the basis of the book value of assets but rather on the income producing ability of those assets, many of which were already fully depreciated.
“I’ll deflate the total assets and long-term debt by the PPI”
I’m not sure what that will tell you. As I see it, the distortion for debt ratios is caused by assets and debt on the books being acquired at different periods. If the book values of assets represent sums paid for those assets many years earlier – and the debt refers to debt acquired relatively recently – then the ratio of debt to assets is distorted. That has been exactly the case at large companies at which I’ve worked. Debt was extended to us not on the basis of the book value of assets but rather on the income producing ability of those assets, many of which were already fully depreciated.
The timing issue is a valid point. I don’t think one can ever get a single “best” measure of the debt ratio for a given company, which is why we use several. As you point out, another measure of debt is the income support the company has, like EBITDA or free cash flow, or EBIT.
Debt/asset and debt/equity ratios should really be used in conjunction with interest support measurements, like times-interest-earned, to measure how leverage has changed. Unfortunately, that’s a level of detail not really feasible for a comment on a blog post.
Furthermore, the data I typically work with is firm-level data, but I’m simply talking in aggregates here. So, yes, on average firms’ leverage has gone down (albeit slightly), as cash holdings (among large firms) have increased. This appears to be in response to higher volatility of operating income. The tax regime hasn’t changed enough to reorder firms’ priorities.
Now, regarding the timing issue: this is important for individual firms. In an aggregate sense, I think the timing issue would tend to wash out, as assets and debt get replaced over time. So the trends in leverage should be reasonably accurate. The leverage (long-term debt/total assets) ratio tends to be 0.18 – 0.2, but the lowest one is 0 and the highest is around 1.3, so there is considerably cross-sectional differences.
The timing issue is a valid point. I don’t think one can ever get a single “best” measure of the debt ratio for a given company, which is why we use several. As you point out, another measure of debt is the income support the company has, like EBITDA or free cash flow, or EBIT.
Debt/asset and debt/equity ratios should really be used in conjunction with interest support measurements, like times-interest-earned, to measure how leverage has changed. Unfortunately, that’s a level of detail not really feasible for a comment on a blog post.
Furthermore, the data I typically work with is firm-level data, but I’m simply talking in aggregates here. So, yes, on average firms’ leverage has gone down (albeit slightly), as cash holdings (among large firms) have increased. This appears to be in response to higher volatility of operating income. The tax regime hasn’t changed enough to reorder firms’ priorities.
Now, regarding the timing issue: this is important for individual firms. In an aggregate sense, I think the timing issue would tend to wash out, as assets and debt get replaced over time. So the trends in leverage should be reasonably accurate. The leverage (long-term debt/total assets) ratio tends to be 0.18 – 0.2, but the lowest one is 0 and the highest is around 1.3, so there is considerably cross-sectional differences.