Designing the Optimal Capital Structure: No Framework, Just Trade-offs

In April 2013 Apple, sitting on roughly $145bn of cash, borrowed $17bn in the bond market. At the time it was the largest corporate bond deal ever done. The reason was not a shortage of money. Most of the cash sat offshore and repatriating it would have triggered US tax, so the company borrowed cheaply in the US to fund dividends and buybacks rather than incur the tax bill. It was a capital structure decision driven by tax, investor expectations and cost.

There is no formula that produces the optimal capital structure. There is a set of trade-offs, and the job is to choose which ones give you the best outcomes. This article looks at eleven of them in turn:

• Cost
• Flexibility
• Dilution
• Leverage
• Liquidity
• Control
• Rating
• Covenants
• Security
• Maturity
• Investor Expectations

Eleven considerations, and not one of them settled by a calculation.

We explore the Cost of Capital in more detail in a separate article; this one is about the decisions beyond that calculation.

Cost and flexibility pull against each other

Debt is cheaper than equity because lenders rank ahead of shareholders and interest is tax-deductible. The catch is that the cheapest debt is the least flexible.

A secured fixed-rate bond with a make-whole call (a prepayment clause that charges you the present value of all the interest the investor would have received) is cheap precisely because you cannot get out of it. A term loan with tight covenants and a full security package is cheap because the lender has boxed you in.

Flexibility, meaning the ability to prepay, to add debt, to sell an asset or to change strategy without asking permission, is a thing you buy with a higher margin. The entire private credit industry has grown up to service CFOs that realise the value of this flexibility.

Dilution and leverage are the same decision seen from two sides

Every euro of growth must be funded by someone.

Fund it with equity and existing owners give up a share of the future: Dilution.

Fund it with debt and they keep the share but add a fixed obligation that must be met whether or not the growth arrives: Leverage.

Founders and private owners tend to fear dilution more than leverage, and the pattern in mid-market businesses is a balance sheet with more debt than the cash flows comfortably support, held in place by a reluctance to let anyone else in. Public companies often run the opposite bias. Leverage is disciplined by the rating and dilution is seen as cheap because the share price is high. Neither instinct is wrong. The mistake is to hold one without having measured the other. The useful question is what the marginal euro of growth costs under each route, in a bad year, not a good one.

Leverage and liquidity are not the same thing

A company can be modestly levered and illiquid, if its debt is short-dated and its revolver is fully drawn. It can be heavily levered and liquid, if it holds cash and long-dated bonds. The Covid outbreak filtered companies by liquidity, not leverage. The ones that drew their revolvers early and held the cash held up well, whatever their leverage ratio said. Liquidity buffers should be built to the stressed scenario, separate from the leverage decision. Undrawn committed lines are a cost of liquidity that is worth paying.

You give up Control either way

Debt takes control through covenants, consent rights and, ultimately, security enforcement.

Equity takes it through voting rights, board seats, liquidation preferences and, in venture deals, a long list of investor consents.

The trade-off is usually framed as debt versus equity, but within each there is a spectrum. A bilateral loan from a relationship bank is a very different control proposition from a unitranche from a private credit syndicate, which will want board observer rights and monthly reporting. A growth-equity minority stake with a drag-along right is a different animal from a public float.

The scrutiny of the public eye

The rating is a constraint you choose

A credit rating turns your capital structure into a public commitment. Once you hold an investment-grade rating, every decision about leverage, buybacks and acquisitions is made with half an eye on the ratings agencies opinion. The benefit is cheaper financing and access to a wider investor base, the cost is flexibility, and the cliff risk of a potential downgrade below investment grade.

Kraft Heinz and Ford both lost investment grade in early 2020, and the spread on their debt widened by multiples, not basis points, because a whole class of investors is not permitted to hold high yield. Companies that sit at BBB minus with a leverage target close to the agency's trigger have chosen the cheapest rating with the most expensive risk.

Investor expectations

Once you have shareholders, or bondholders, each has a view about how the company should be managed, and disappointing them has a cost. Shareholders who were promised a progressive dividend will punish a cut, even a sensible one. Bond investors who bought on the basis of a stated leverage range will sell if you breach it for an acquisition. A private equity sponsor expects leverage to be used, because that is how the return is made. Venture investors push for massive growth to compensate, trying to find the 100x winners that make up for the bets that didn't get off the ground.

At a certain level, managing these expectations requires dedicated resources. We cover Investor Relations in more detail in a separate article.

Covenants and security are the price of lower margins

• A covenant is a restriction in the loan agreement, usually financial (leverage below a level, interest cover above one) and tested quarterly.
• Security is a charge over assets the lender can enforce if the promises fail.

Both tools reduce risk for the lender, and therefore allow them to accept a lower margin in exchange for the ability to intervene earlier and recover more. The questions are how much of each you can tolerate and where.

A maintenance covenant set at 3.0 times when you run at 2.5 might seem comfortable, against your forecasts, but it could be a serious imposition in a down-case. Security granted to a bank today is security you cannot offer a bond investor tomorrow, and a negative pledge, a promise not to grant security to anyone else, in an unsecured bond can make a future secured bank facility impossible without a consent process.

Capital structures are path-dependent. What you grant now can constrain what you can do next.

Maturity is where structures actually break

Long-dated money costs more, in a higher spread and in the risk of being locked into a rate you do not like if rates fall.

Short-dated money is cheaper but must be replaced more often, at whatever the market is charging that year.

The answer is a debt ladder, with no single year carrying more than the business could refinance in a difficult market, and the average maturity matched to the assets being funded.

Other considerations

How the eleven connect

Some of the eleven pull directly against each other, some are the same decision seen from a different angle, and some are simply confused with one another. It helps to know which is which before you start trading them off.

Four kinds of relationship: what pulls against what, what is the same decision twice, what gets confused, and what comes from outside.

Putting it together

There is no scorecard that resolves eleven trade-offs into one number. What works in practice is a short documented capital structure policy.

• Target leverage range
• Minimum liquidity in months of stressed obligations
• Maximum share of debt maturing in any one year
• Fixed-to-floating rate target
• A target rating or covenant headroom you are committed to protecting

Four or five lines, agreed, reviewed periodically. Every major financing decision should be checked against them.

The same job at three sizes