Chapter 25: Surety Bonds: Underwriting the Promise to Perform

"A surety bond is not insurance against the contractor's failure. It is a banker's bet that the contractor will succeed — written by an underwriter who fully expects to be paid back if they are wrong." — Constructed teaching line, in the voice of this book. It compresses the chapter's whole argument: in surety, the underwriter is a lender in everything but name, and "expects to be paid back" is not a figure of speech.

Overview

On your desk is a request that looks, at first glance, like every other commercial submission you have worked this part: a contractor, a set of financials, a broker who wants terms. But read the cover note again. The contractor is not asking you to insure its building or its trucks or its liability. It is asking you to stand behind a promise it has made to someone else — to guarantee, to a project owner it has never worked for, that it will finish a \$6 million water-treatment plant on time, to specification, and pay every subcontractor and supplier along the way. If the contractor fails, the project owner will turn to you. And here is the part that should make you sit up: when you pay that owner, you will then turn around and try to collect every dollar back from the contractor and its owners personally, because they signed an agreement promising exactly that. You are not insuring a risk. You are extending credit.

That single fact — that surety is credit dressed in the clothing of insurance — reorganizes everything you learned in the first twenty-four chapters. The loss ratio you have been trained to manage is supposed to run near zero. The "premium" is not a premium at all; it is closer to a fee for a guarantee, priced like a service, not like a pool of expected losses. The adverse-selection problem you have fought since Chapter 1 takes a different shape, because the worst risks are not the ones who want coverage most — they are the ones whose books look fine right up until the job runs out of cash. And the most important document in the file is not a loss run or an inspection report. It is a financial statement, read the way a careful commercial banker reads it, looking for the working capital that will carry a contractor through the lean middle of a job.

This chapter teaches you to make that credit decision. We start with the structural difference that defines the whole line — the three-party relationship that makes surety unlike any two-party insurance you have seen. We tour the two great families of bonds: contract surety, which guarantees construction and service obligations, and commercial surety, which guarantees a sprawling miscellany of legal and regulatory promises. We reframe the underwriting as lending and meet the discipline at its heart, the three C's of character, capacity, and capital. We read a contractor's financials the way a surety must. We weigh the soft, decisive factor of the contractor's character and the hard limit of its capacity. And we follow a bond all the way to the worst case — a default — to see why a surety loss is recovered, fought over, and salvaged in a way an insurance loss never is.

In this chapter, you will learn to:

• Explain the three-party relationship at the core of every surety bond and why it makes surety a guarantee of performance, not a transfer of risk.
• Distinguish the major contract surety bonds (bid, performance, payment) and commercial surety bonds (license, court, fiduciary), and say what each one guarantees and to whom.
• Underwrite a surety account as a credit decision using the three C's, and explain why a surety underwrites toward zero expected loss.
• Read a contractor's financial statement for working capital, net worth, backlog, and the bank line.
• Explain what happens when a bond is called — the surety's options, indemnity, and salvage — and why a surety loss behaves nothing like an insurance loss.

Learning Paths

This chapter sits in the commercial-lines heart of the book, but it teaches a way of thinking that reaches well beyond construction. Read it as the chapter where "underwriting" and "lending" turn out to be the same craft seen from two sides.

🏠 Personal Lines: Surety is almost entirely a commercial discipline, but the logic — guaranteeing a third party against someone's failure to perform, then seeking repayment — appears in personal life as the co-signer on a loan and the bail bond. Read §25.1 and §25.4 for the mental model; the financial-analysis sections are optional for you. 🏢 Commercial Lines: This is your chapter. Surety is a major commercial line with its own market, forms, and underwriters. The contract-surety sections (§25.2, §25.5, §25.6, §25.7) are core; note how differently this account is judged from the Harbor Steel property and liability you have been building. 📊 Analytics: Surety is the line where data does the least and judgment does the most — expected loss is near zero, so frequency/severity modeling barely applies. Watch §25.4 and §25.7 for why a low-frequency, high-severity, credit-driven line resists the modeling that prices auto and property. 📜 Certification: Surety appears in the AINS, AU, and CPCU bodies of knowledge as a distinct line with its own vocabulary. The key terms here — the three parties, the bond types, the three C's — are exam staples; §25.1 through §25.4 carry the testable core.

25.1 How surety differs from insurance: the three-party relationship

Begin with the structure, because in surety the structure is the subject. Every insurance contract you have studied so far is a two-party arrangement: an insurer and an insured, with the insurer promising to pay the insured (or, in liability, to pay on the insured's behalf) when a covered loss strikes. The money flows in one direction when things go wrong, from insurer to insured, and the insurer never expects to get it back. That is the whole shape of insurance, and you have internalized it.

A surety bond breaks that shape. A surety bond is a three-party contract in which one party (the surety) guarantees to a second party (the obligee) that a third party (the principal) will fulfill an obligation — and if the principal fails, the surety will make the obligee whole, then seek reimbursement from the principal. Read that sentence twice, because three features hide inside it and each one overturns an instinct insurance gave you.

The first feature is the three-party relationship itself, the term this chapter owns. The three parties are the principal (the party who must perform the obligation — the contractor, the licensed plumber, the estate administrator), the obligee (the party protected by the bond, who receives the obligation — the project owner, the state licensing board, the probate court), and the surety (the party guaranteeing the principal's performance to the obligee — your company). The obligee is the beneficiary; the principal is the one whose performance is guaranteed; the surety stands behind the principal. In insurance, the party who pays the premium is the party the insurer protects. In surety, the principal pays the premium, but the obligee is protected. That split — the payer is not the protected party — is the single most disorienting thing for an insurance professional crossing into surety, and you should fix it in your mind now.

THE TWO STRUCTURES, SIDE BY SIDE                              [constructed teaching diagram]

  INSURANCE (two parties)              SURETY (three parties)

   INSURED  ──premium──►  INSURER       PRINCIPAL ──premium──► SURETY
      ▲                      │            (contractor)            │
      │                      │                  │                 │ guarantees
      └──pays the loss───────┘                  │ owes the        │ performance
        (no expectation                         │ obligation      ▼
         of repayment)                           ──────────►  OBLIGEE
                                                  performs        (project owner)
                                                                    ▲
                                          if principal fails, surety pays
                                          obligee ──► then RECOVERS from principal

The second feature follows from the first: the surety expects to be reimbursed. When a property insurer pays a fire claim, the matter is closed; the insurer absorbs the loss as the cost of doing business and the premium pool funds it. When a surety pays an obligee because the principal defaulted, the surety turns immediately to the principal — and, because of an agreement everyone signed at the outset (the indemnity agreement, which we develop in §25.7), to the principal's owners personally — and demands its money back. The bond is, in legal effect, a guarantee, the same instrument as a co-signature on a loan. The economic substance is credit: the surety lends its financial strength and its name, betting that the principal will perform and that, in the rare case it does not, the principal will be good for the loss.

The third feature is the one that should change how you price: a surety underwrites toward zero expected loss. Because the surety expects reimbursement, and because it only bonds principals it judges able to perform, the line is built on the premise that losses should be rare and largely recoverable. This is not a pious hope; it is the business model. The "premium" — properly the bond premium or bond fee — is not the price of an expected loss the way an auto premium is. It is a charge for the use of the surety's credit and capacity, set as a rate per thousand dollars of the bond's penal sum, and it is small precisely because the surety does not expect to pay. Where a property line might run an 60% loss ratio in a normal year and still profit, a surety line that runs a 60% loss ratio is a catastrophe; the whole account structure assumes a loss ratio in the low double digits or less across the cycle, with the bulk of paid losses eventually clawed back.

📋 At the Desk The cleanest test of whether you are looking at surety or insurance is to ask: after a loss, does the party who paid expect to get it back? If no, it is insurance — the insurer pools the premium and absorbs the loss. If yes, it is suretyship — the surety guaranteed someone else's promise and will pursue them for reimbursement. Everything downstream follows from that answer. It tells you the loss ratio should run near zero, that the file's center of gravity is a credit analysis rather than a hazard analysis, that the principal's balance sheet matters more than its loss runs, and that the worst-case scenario is not a covered peril but a business failure. Underwriters who come to surety from property or casualty keep reaching for the wrong tools for a year; the ones who succeed retool fast around this single question.

There is a fourth, quieter point worth making while the structure is fresh, because it explains why surety exists at all. The obligee — the project owner who requires a performance bond, the state that requires a contractor's license bond — uses the bond as a prequalification device as much as a guarantee. When a public agency requires a \$6 million performance bond on a \$6 million job, it is outsourcing a credit and competence check to the surety industry. A contractor who can obtain the bond has been vetted by a surety that put its own money behind the judgment; a contractor who cannot has, in effect, failed that vetting. The bond protects the obligee against the principal's default, yes — but its larger social function is to keep unqualified or undercapitalized contractors off serious work in the first place. You are not only a guarantor. You are a gatekeeper, and the gate you guard protects the public's money and the subcontractors' livelihoods. That is theme six — insurance serves a social function — wearing a hard hat.

25.2 Contract surety: bid, performance, and payment bonds

The largest and most important branch of the line is contract surety: bonds that guarantee a contractor's obligations on a construction or service contract. Most of these bonds exist because the law requires them. On U.S. federal construction projects above a statutory threshold, the Miller Act requires performance and payment bonds; the states have their own "Little Miller Acts" imposing parallel requirements on state and local public work. So a vast share of public construction in America cannot proceed unless a surety has agreed to stand behind the contractor — which makes the surety industry a silent precondition of the nation's roads, schools, and water plants. Contract surety comes in three core forms, and they work in sequence across the life of a project.

The bid bond comes first, at the moment a contractor submits a bid. It guarantees that if the contractor is awarded the contract, it will actually enter into the contract at the bid price and furnish the required performance and payment bonds. The risk it protects against is the contractor who bids low — perhaps by mistake, perhaps to win and then renegotiate — and then walks away when the number proves unworkable, forcing the owner to award to the next-highest bidder at greater cost. The bid bond's penal sum is typically a percentage of the bid (often 5% to 10%), and it pays the difference between the defaulting contractor's bid and the cost of the next-acceptable bid, up to that penal sum. A bid bond is the surety's statement that it has already pre-qualified this contractor to take the job — because the surety that issues the bid bond is, in practice, committing to issue the performance and payment bonds if the bid wins.

The performance bond is the heart of contract surety. It guarantees that the contractor will perform the contract according to its terms — finish the work, on schedule, to specification. If the contractor defaults, the surety must, at its option, complete the work itself, finance the original contractor to completion, arrange for a replacement contractor (a tender), or pay the obligee its damages up to the bond's penal sum — the maximum dollar amount of the surety's liability, usually set at 100% of the contract price. The performance bond is what makes the project owner whole if the contractor goes under mid-job, and it is the bond whose call costs a surety the most, because completing a half-built, mismanaged, possibly defective project is enormously expensive — often more than the remaining contract funds, sometimes far more.

The payment bond runs alongside the performance bond and guarantees something the owner cares about for a specific legal reason: that the contractor will pay its subcontractors, laborers, and material suppliers. Why does the owner require a bond protecting the subcontractors? Because on private work, an unpaid subcontractor can file a mechanic's lien against the owner's property, and on public work — where public property generally cannot be liened — the payment bond is the subcontractors' only recourse. The payment bond lets the owner ensure that everyone downstream gets paid, so that liens do not cloud the title and subcontractors are not ruined by a contractor who collected from the owner but failed to pay its trades. The payment bond's penal sum is also commonly 100% of the contract price, and on a defaulting job a surety frequently pays under both the performance and the payment bond at once — completing the work and paying the unpaid suppliers — which is why a single contractor failure can generate a loss far larger than any one bond's penal sum suggests.

THE CONTRACT-SURETY SEQUENCE ON ONE PROJECT            [constructed teaching example]

  BID STAGE          AWARD                CONSTRUCTION                COMPLETION
  ─────────          ─────                ────────────                ──────────
  BID BOND      ──►  contractor signs ──► PERFORMANCE BOND   +  PAYMENT BOND   ──► bonds
  guarantees the     the contract at      guarantees the work    guarantees the      released
  contractor will    the bid price &      is completed per       subs & suppliers    when work
  honor its bid &    furnishes the        the contract           are paid            is accepted
  post the bonds     P&P bonds            (penal sum ~100%)       (penal sum ~100%)   & no claims

  Penal sum:  ~5–10% of bid    100% of contract        100% of contract

⚖️ Compliance Corner Contract surety is one of the few corners of insurance whose existence is mandated by statute rather than chosen by the buyer. The federal Miller Act and the state Little Miller Acts are the real legal authority here, and they matter to you in a specific way: on bonded public work, the form of the bond is often prescribed by the statute or the contracting agency, not negotiated. You frequently cannot add the exclusions or sublimits you would reach for in a liability policy, because the obligee's statute dictates the wording. That rigidity is the price of playing in public construction, and it means your underwriting protection comes almost entirely from whom you bond and on what terms you indemnify — the credit decision and the indemnity agreement — rather than from policy language. When you cannot tailor the contract, you must tailor the selection.

A word on capacity before we leave contract surety, because it frames the financial analysis to come. A surety does not bond a contractor for one job at a time in isolation; it establishes a bonding program for the contractor — a single-job limit (the largest project it will bond) and an aggregate limit (the total of all bonded work outstanding at once). These limits are the surety's judgment about how much work the contractor can safely carry, and they are the real product the contractor is buying: not a single bond but a line of bonding capacity it can draw against as it bids new work. Setting that program is the central act of contract-surety underwriting, and §25.5 and §25.6 are about how you set it.

25.3 Commercial surety: license, court, and fiduciary bonds

Alongside contract surety runs a second, more miscellaneous family: commercial surety, the broad category of bonds that guarantee obligations other than construction or service contracts. If contract surety is about building things, commercial surety is about a thousand smaller promises that businesses and individuals must guarantee to governments, courts, and other parties in order to operate. The category is sprawling — there are hundreds of bond types — but it organizes naturally into three groups.

License and permit bonds guarantee that a licensed business will comply with the laws and regulations governing its activity, and will pay any penalties, taxes, or damages its non-compliance causes. A contractor's license bond, a motor-vehicle dealer bond, a mortgage-broker bond, a freight-broker bond, a plumber's or electrician's license bond — all of these are required by a government agency as a condition of issuing a license, and all protect the public (or the agency) against the licensee's misconduct or failure to pay what it owes. The obligee is the licensing authority; the beneficiaries are the members of the public the regulation protects. These bonds are typically modest in penal sum, high in volume, and the bread-and-butter of commercial surety. They behave a little more like insurance than contract bonds do, because the underwriting is lighter and more automated, but the structure is still a three-party guarantee with reimbursement, never a pooled transfer of loss.

Court bonds are required by a court as a condition of taking some action in litigation, and they split into two sub-families. Judicial bonds protect a party against loss from a court proceeding — an appeal bond (or supersedeas bond), for instance, guarantees that a party appealing a money judgment will pay it (plus interest and costs) if the appeal fails, so the winning party is protected during the delay; an attachment or injunction bond protects a defendant against wrongful seizure or restraint. Fiduciary bonds (sometimes called probate bonds) guarantee the faithful performance of someone appointed by a court to manage assets for others — an executor or administrator of an estate, a guardian, a trustee, a conservator. The obligee is the court, and the beneficiaries are the heirs, wards, or creditors whose money the fiduciary controls. A fiduciary bond guarantees that the executor will not embezzle the estate, will account honestly, and will distribute the assets as the law and the will require.

Public official and miscellaneous bonds round out the family: bonds guaranteeing the honest performance of an elected or appointed public official (a treasurer, a tax collector, a notary), and a long tail of specialized bonds — customs bonds, reclamation bonds for mining, utility-deposit bonds, lost-instrument bonds, and more. Each guarantees a particular promise to a particular obligee, and each is underwritten according to the credit and character of the principal and the size and nature of the obligation.

THE COMMERCIAL-SURETY MAP                                   [constructed teaching example]

  LICENSE & PERMIT          COURT                        PUBLIC OFFICIAL & MISC.
  ────────────────          ─────                        ───────────────────────
  contractor license        JUDICIAL                     public official (treasurer,
  motor-vehicle dealer        appeal / supersedeas         tax collector, notary)
  mortgage / freight broker   attachment / injunction     customs / import
  plumber / electrician     FIDUCIARY (probate)          reclamation (mining)
  ── obligee: a govt          executor / administrator    utility deposit
     licensing agency         guardian / trustee          lost instrument
  ── protects: the public     conservator                 ── many hundreds of types
                            ── obligee: the court

⚠️ Underwriting Trap The trap in commercial surety is to treat a small penal sum as a small risk and wave the bond through on the penal sum alone. Two things make that dangerous. First, some commercial bonds are cumulative or have a long tail: a license bond may remain in force for years and respond to violations across that whole period, so a \$25,000 penal sum can be exposed many times over the bond's life if the obligee's statute allows multiple claimants or annual reinstatement. Second, a fiduciary bond's penal sum may be modest relative to the assets the fiduciary controls — and the temptation to defalcate scales with those assets, not with the bond. A dishonest executor sitting on a large estate is a real exposure even under a small bond, and the bond can be exhausted by the first claimant, leaving later heirs unprotected. The disciplined commercial-surety underwriter sizes the exposure, not just the penal sum, and reads the obligee's underlying statute to learn how the bond can actually be called.

It is worth noting how commercial surety has industrialized at the small end. License-and-permit bonds in modest amounts are increasingly written through automated programs and agency systems — the principal answers a short application, a credit check runs, and a bond issues in minutes, much like the straight-through processing you met in small commercial (Chapter 20 owns that idea). This works for the same reason small-commercial automation works: the bonds are homogeneous, the amounts are small, the credit signal is largely captured by a score, and the volume justifies the build. But the moment the penal sum rises, or the bond type carries a meaningful default history, or the principal's credit is thin, the file goes to a human surety underwriter who reads it as a credit. The line between the automated tail and the judged account is, once again, the line between simple, high-volume risk and complex, individual risk that runs through the whole book (theme five — technology augments, it does not replace).

25.4 Underwriting surety as lending: the three C's

We can now state the discipline directly: to underwrite surety is to make a credit decision. The question is not "what is the probability and severity of a loss, and what premium covers it?" — the insurance question. The question is "will this principal perform its obligation, and if it somehow cannot, is it good for the money?" — the lender's question. A surety underwriter is, functionally, a commercial credit analyst who happens to issue a guarantee instead of a loan. The whole apparatus of credit analysis applies, and the industry has compressed it, as bankers long have, into the three C's: character, capacity, and capital. Master these and you have the spine of every surety decision.

Character is the principal's integrity, reputation, and track record — the willingness to perform and to deal honestly. In surety, character is not a soft nicety; it is frequently the decisive factor, because a contractor of genuine integrity will fight to finish a troubled job and will work with the surety, while a contractor of poor character will walk, hide assets, or fight the surety in court. Character is assessed through the contractor's history of completed jobs, its reputation among owners, architects, and subcontractors, its payment record with suppliers, the longevity and stability of the business, any history of liens, judgments, or disputes, and the surety underwriter's own read of the people across the table. Many seasoned surety underwriters will tell you that they have lost money on contractors with strong financials and weak character, and rarely on the reverse.

Capacity is the principal's ability to perform the obligation — the technical, managerial, and operational competence to do the work it is bonded for. For a contractor, capacity means the right trade skills and equipment, experienced project managers and field supervisors, a track record on jobs of the size and type now contemplated, and the organizational depth to run multiple projects at once without the wheels coming off. A capable mason is not automatically capable of building a hospital; a contractor that runs three \$2 million jobs smoothly may not be able to run one \$15 million job. Capacity is why a surety sets a single-job limit and an aggregate limit: it is fencing the contractor inside the size and volume of work the surety believes it can actually execute. The most common cause of contractor default is not dishonesty but overreach — a competent firm that took on a job too big, too unfamiliar, or too far from home, and could not manage it. Underwriting capacity is largely the discipline of keeping good contractors from overreaching.

Capital is the principal's financial strength — the working capital, net worth, liquidity, and access to credit that let it absorb the inevitable cash-flow strains of its work and survive a bad job. Capital is where the surety reads the financial statements (§25.5), and it is the C that most resembles ordinary credit analysis. But note the order in which surety underwriters traditionally weight the three: many will say character first, capacity second, capital third — that capital, while necessary, cannot save a job that a dishonest or incompetent contractor is botching, whereas a contractor of strong character and capacity will often find a way through a thin-capital season. Capital is the floor you must clear; character and capacity are what decide the account once the floor is cleared.

THE THREE C's — AND WHAT EACH ANSWERS               [constructed teaching example]

  CHARACTER   Will they perform honestly and stand behind the job?
              ── reputation, track record, payment history, liens/judgments, the people
  CAPACITY    Can they actually do THIS work, at THIS size, at THIS volume?
              ── trade skill, equipment, management depth, experience on like jobs
  CAPITAL     Can they absorb cash strain and survive a bad job?
              ── working capital, net worth, liquidity, bank line, profit history

  Traditional weighting in contract surety:  CHARACTER  >  CAPACITY  >  CAPITAL
  (capital is the floor you must clear; character and capacity decide the account)

🤖 Model vs. Judgment Surety is the line where the algorithm has the least to work with, and it is worth understanding why. The models that price auto and property feed on large pools of homogeneous risks with frequent, observable losses — exactly what frequency/severity modeling needs. Surety has the opposite data: losses are rare (by design), each large loss is idiosyncratic, and the decisive variable, character, is precisely the one that resists quantification. Credit scores and financial ratios genuinely help at the small, high-volume end of commercial surety, and analytics can flag a deteriorating contractor balance sheet faster than a human reviewing statements annually. But the core contract-surety decision — will this contractor, on this job, perform — turns on a read of people, plans, and circumstances that no current model captures. This is the one line in the book where "the underwriter's judgment is irreplaceable" is not an aspiration but a description. The future surely brings better data on contractor performance; it is unlikely to bring a model that can sit across a table and decide whether to trust someone.

There is one more reframing the three C's force on you, and it returns us to a theme from Chapter 1. Adverse selection (Chapter 1 owns the term) operates in surety, but with a twist. In insurance, the bad risk is the one most eager to buy. In surety, the bad risk is often the contractor whose financials look adequate but whose backlog is growing faster than its capital, who is bidding bigger jobs to outrun a cash-flow problem, who needs the bond not to win good work but to keep a teetering operation moving. The surety underwriter's defense against this is not a higher price — raising the bond fee does nothing to offset a default that wipes out years of fees in one job. The defense is selection and capacity control: declining the overreaching contractor, or capping its program below the level at which it can hurt you. Pricing does not follow risk in surety the way it does in insurance (theme four bends here), because you cannot price for a loss you must not allow to happen. You manage the risk by not writing it, or by limiting how much of it you write.

25.5 Financial analysis: working capital, net worth, backlog, and the bank line

Capital is the C you can measure, and the measurement happens in the contractor's financial statements. A surety underwriter reads those statements the way a commercial banker does — looking past the revenue and the profit at the liquidity and the staying power that determine whether a contractor can survive the cash-flow valleys of its work. Four figures matter most, and you should know what each tells you and what it does not.

Working capital — current assets minus current liabilities — is the single most-watched number in contract surety. It measures the contractor's short-term liquidity: the cushion of cash, receivables, and other near-cash assets available to fund operations after the near-term bills are paid. Working capital matters so much because construction is a brutally cash-hungry business. A contractor must pay for labor, materials, and equipment before it gets paid by the owner, often with a retainage (a percentage of each progress payment) held back until the job is complete. A contractor can be profitable on paper and still fail because it ran out of cash in the lean middle of a big job. A common rule of thumb in the industry is that a contractor's bonding program is sized as a multiple of its working capital — you may hear figures like ten or fifteen times working capital as a rough aggregate ceiling — though every surety sets its own, and the multiple is a starting point a judgment then adjusts, never a formula that decides the account.

📋 At the Desk When you read a contractor's balance sheet for working capital, you do not take the current-asset figure at face value — you quality-adjust it, exactly as a careful lender would. Underbillings (costs and estimated earnings in excess of billings) may not be as liquid as cash; aged receivables past ninety days are discounted or thrown out; receivables from a single shaky owner are scrutinized; inventory and work-in-process specific to one job may be worth little if that job dies; and related-party receivables (money the owner's other company owes the contractor) are often disallowed entirely. The headline working capital and the adjusted working capital can differ enormously, and the gap is itself a signal — a contractor whose liquidity depends on soft assets is a contractor whose cushion may not be there when a job goes wrong. Read the notes to the financial statements as carefully as the statements themselves.

Net worth — total assets minus total liabilities, the owner's equity in the business — measures the contractor's overall financial substance and its capacity to absorb a serious loss. Where working capital is about short-term survival, net worth is about long-term staying power and the depth of the cushion behind the surety's guarantee. A surety also reads net worth for its quality: equity built from retained earnings over years of profitable work is worth more than equity inflated by revalued real estate or goodwill, and a net worth heavily tied up in illiquid or related-party assets offers thin real protection. Net worth also anchors the single-job limit in a way working capital anchors the aggregate, though again the relationship is a guide for judgment, not a fixed formula.

Backlog — the total value of work the contractor has under contract but not yet completed — measures how much the contractor has already committed to deliver, and it is read against capacity and capital. A healthy backlog shows a contractor with steady future work; an excessive backlog is one of the loudest warning signs in surety, because it means the contractor has taken on more work than its capital and management can safely carry. The surety watches the relationship between backlog, working capital, and the contractor's historical volume: a firm that has always done \$20 million a year and suddenly carries \$60 million in backlog has either grown explosively (which strains everything) or overreached (which kills contractors). Backlog is also read for quality — the margins built into the jobs, the reputation of the owners, the presence of any single job large enough to sink the firm, and whether the work is the kind the contractor has done before.

The bank line — the contractor's line of credit and its relationship with its bank — measures the contractor's access to liquidity beyond its own balance sheet, and it tells the surety something it cannot fully see in the statements: what another sophisticated credit grantor thinks of this contractor. A strong, long-standing bank relationship with an unused or lightly used line is a powerful positive; it means the contractor has a liquidity backstop for a bad job and that its bank, which has its own credit analysts and its own money at risk, has judged the firm creditworthy. A contractor that is maxed out on its line, or that has no real bank relationship, has no cushion behind its working capital and no independent vote of confidence — and the surety is being asked to be the only credit grantor standing behind the firm, which is exactly the position a surety does not want to occupy alone.

📄 Read the Submission

text FIGURE 25.1 — "Two contractors, one bond request" [constructed teaching example] THE SUBMISSION Two general contractors each request a $5M single-job / $12M aggregate bonding program to bid public water-treatment work. Same trade, same region, same bond fee quoted. THE CONTEXT Contractor A: $1.4M adjusted working capital, $3.2M net worth, $9M backlog (steady at its historical ~$18M/yr volume), an unused $2M bank line, 22 years in business, clean payment record. Contractor B: $1.5M reported working capital (but ~$0.6M of it is a receivable from the owner's other company), $2.0M net worth, $14M backlog after years around $8M/yr, a fully-drawn $1M bank line, 4 years in business, two supplier liens resolved late. WHAT IT SHOWS A looks like a fundable program: liquidity, equity, steady backlog, an untapped line, a long clean record. B shows the classic overreach pattern — backlog nearly double its historical volume, soft working capital propped by a related-party receivable, no liquidity backstop, thin and recent track record, late payments under stress. WHAT IT DOESN'T The statements alone don't prove B will fail or A will succeed; they don't show the quality of B's management or whether its growth is a genuine opportunity. The job references, the work-in-process schedule, and a meeting with the principals would. THE DECISION Approve A's program as requested. For B: decline the program as requested; if anything, offer a much smaller program (a single job well within its demonstrated volume) and require stronger indemnity — but a decline is defensible. You cannot price your way out of B's overreach. THE LESSON In surety you underwrite the contractor's capacity to carry the WORK, not just the penal sum of one bond. The danger is the good-looking firm growing faster than its capital — and the cure is selection and program limits, not a higher fee.

A note on the form of the financial statements, because it carries underwriting weight. Contractors present financials at three levels of accountant involvement: compiled (the accountant assembles the numbers management provides, with no assurance), reviewed (limited analytical assurance), and audited (the highest assurance, with the accountant independently verifying). For a small bonding program a compiled or reviewed statement may suffice; as the program size rises, sureties require reviewed or audited statements, and they strongly prefer statements prepared on the percentage-of-completion method by a construction-experienced CPA, because that method shows the real earnings and the over/underbillings on each job. A contractor that cannot or will not produce statements at the quality its program size demands has told you something — and the work-in-process schedule (the job-by-job listing of contract values, costs to date, estimated costs to complete, and billings) is often the single most revealing page in the file, because it shows whether the contractor is making or losing money on the jobs it is currently running, before the annual statement catches up.

25.6 The contractor's character and capacity

Numbers tell you whether a contractor can survive a bad job. They do not tell you whether it will finish a good one. For that you return to the two C's that resist the spreadsheet — character and capacity — and here the surety underwriter's craft looks less like banking and more like the judgment this whole book is about. The financials are necessary; they are rarely sufficient. The accounts that lose sureties money most often passed the financial screen and failed on character or capacity, which is why experienced surety underwriters invest so heavily in the parts of the file a model cannot read.

Assessing character is, in large part, reference work and pattern reading. You talk to the owners and architects the contractor has worked for: did it finish on time, deal fairly with change orders, stand behind defective work, treat the owner as a partner or an adversary? You talk to the contractor's subcontractors and suppliers: does it pay on time, or does it stretch its trades to finance its own cash flow? (A contractor that habitually pays its subs late is showing you both a character signal and a liquidity signal at once.) You read the public record for liens, judgments, and litigation, and you weigh not just their presence but how the contractor handled them. And you read the people — the owner's candor about problems, the depth and stability of the management team, whether the second generation taking over a family firm has the judgment the founder had. None of this is quantifiable, and all of it is load-bearing.

Assessing capacity is the discipline of matching the contractor to the work. You ask whether this contractor has actually built jobs of this size — a firm whose largest completed project is \$4 million is a different risk on a \$5 million job than on a \$20 million one, regardless of what its balance sheet allows. You ask whether it has built this type of work — a contractor expert in commercial buildings may be out of its depth on a water-treatment plant with complex mechanical and process systems. You ask about geography — a contractor performing far from its home base loses its supervisory grip, its supplier relationships, and its labor pool, and "going on the road" is a recurring theme in contractor failures. And you ask about organizational depth: who runs the jobs when the owner is stretched across three projects, and does the firm have the project managers, estimators, and field supervisors to staff its backlog without overloading anyone. Capacity failures are usually not single dramatic mistakes; they are the slow loss of control that comes when a competent firm spreads itself past the limit of its management.

⚠️ Underwriting Trap The most expensive mistake in contract surety is bonding the contractor's ambition instead of its track record — extending a program for the jobs the contractor wants to win rather than the jobs it has proven it can complete. It happens because the pressures all push one way: the contractor wants a bigger program, the agent earns more on bigger bonds, the soft-market competitor will offer it if you don't, and the contractor's last few bigger jobs went fine. The disciplined surety underwriter holds the line on the single-job and aggregate limits, raising them incrementally as the contractor demonstrates it can perform at each new level, and treats a sudden leap in requested capacity as a reason to slow down, not speed up. The contractor that "needs" a program well beyond its demonstrated capacity is often a contractor trying to outrun a problem — and the bond you write to help it grow is the bond that defaults.

The two C's converge in a single practical question the surety underwriter is always really asking: if this job goes wrong, what will this contractor do? A contractor of strong character and adequate capacity will call the surety early, open its books, and work to complete the job and minimize the loss. A contractor of weak character will conceal the problem until it is unsalvageable, divert funds, and leave the surety to discover a disaster too late. This is why the relationship between a surety and its contractors is so much closer and more ongoing than the relationship between a property insurer and its insureds: the surety is extending a continuing line of credit, it reviews the contractor's financials and work-in-process regularly (quarterly or even monthly for larger programs), and it is, in effect, a financial partner watching the contractor's whole operation across years. You are not underwriting a transaction. You are underwriting a relationship — and relationships are judged on character and competence, not on a single year's ratios.

🔍 Check Your Understanding 1. A contractor with strong financials (ample working capital, high net worth, an unused bank line) requests a program well above the size of any job it has ever completed. Which of the three C's is the binding constraint here, and what should the underwriter do? 2. Why does a surety review a contractor's financials and work-in-process several times a year, while a property insurer typically looks at an account once a year at renewal? Tie your answer to the credit nature of the line.

25.7 When surety goes wrong: defaults, indemnity, and salvage

We end where every credit line eventually must be tested: the default. A bond is called when the obligee declares the principal in default and demands that the surety perform under the bond. Most bonds, most years, are never called — that is the whole point of the line. But when a call comes on a contract bond, the surety enters a process that looks nothing like paying an insurance claim, and understanding it tells you why surety is credit all the way down.

When a performance bond is called, the surety does not simply write the obligee a check. It investigates — often before the formal default, because a watchful surety sees the trouble coming in the work-in-process reports — and then chooses among its options. It may finance the original contractor to completion (lending it the money to finish, if the problem is liquidity rather than competence and the contractor is worth saving). It may takeover and complete the work itself, hiring a completion contractor under the surety's own control. It may tender a replacement contractor to the obligee to finish under a new contract. Or, if completion is impractical, it may pay the obligee its damages up to the penal sum and let the obligee complete on its own. Each option has different costs and risks, and choosing among them under time pressure, with a half-finished project deteriorating and an angry obligee, is one of the hardest jobs in the line. The surety's loss is the cost of completion plus any payment-bond claims from unpaid subs minus the contract funds still held by the owner and any salvage — and that net loss can dwarf the bond fee that was ever collected.

This is where indemnity becomes the spine of the whole business. Before a surety writes a single bond for a contractor, it requires a General Indemnity Agreement (GIA) — a contract under which the principal, and typically its owners personally and their spouses, agree to indemnify the surety for any loss, cost, and expense it incurs under any bond. The GIA is what makes surety credit rather than insurance: it is the legal mechanism by which the surety, having paid an obligee, reaches back to the contractor and its owners' personal assets — their homes, their savings, their other businesses — to be reimbursed. The GIA typically also grants the surety powerful rights the moment trouble appears: the right to demand collateral, the right to take control of the contract funds, the right to access the contractor's books and even to run its bonded jobs. A surety underwriter who waives or weakens the indemnity is giving away the core protection of the line, and the strength of the indemnity (who signs it, what assets stand behind it) is a central underwriting decision, not a formality.

ANATOMY OF A PERFORMANCE-BOND LOSS                        [constructed teaching example]

  CONTRACTOR DEFAULTS on a $6M job, ~60% complete, contract funds remaining ~$2.4M
  ────────────────────────────────────────────────────────────────────────────────
  Surety's likely cost to COMPLETE the work          ████████████████  ~$3.6M
  + PAYMENT-BOND claims (unpaid subs/suppliers)       ██████            ~$1.1M
  ──────────────────────────────────────────────────────────────────────────────
  Gross exposure                                                        ~$4.7M
  − remaining CONTRACT FUNDS held by the owner        ██████            ~$2.4M
  − SALVAGE recovered under the GIA (indemnitors,                       ~$0.6M
    collateral, asset recovery)                        ███
  ──────────────────────────────────────────────────────────────────────────────
  = NET SURETY LOSS                                                     ~$1.7M

  Compare: the bond FEE ever collected on this program might have been ~$60–90K.
  One default can erase years of fees — which is why the loss ratio MUST run near zero.

The diagram makes the line's economics visceral. The surety's gross exposure on a single defaulted job runs to millions; the remaining contract funds and the salvage recovered through the indemnity agreement reduce it, but the net loss still dwarfs the modest fees collected across the whole bonding program. Salvage — the surety's recovery of its loss through the contract funds, the indemnitors' assets, collateral, and the pursuit of any responsible parties — is therefore not an afterthought as it might be in property insurance; it is integral to the surety's economics, and a surety with a strong claims-and-salvage operation recovers a meaningful share of its gross losses. This is the final proof that surety is lending: the surety that suffers a loss behaves exactly like a lender whose borrower defaulted, working the collateral and the guarantees to get its money back.

🤖 Model vs. Judgment Notice what the default process reveals about why this line resists automation end-to-end. The decision at a default — finance, complete, tender, or pay — depends on a real-time read of the specific contractor, the specific job, the obligee's temperament, the completion market, and the indemnitors' assets. No model makes that call; an experienced surety claims professional does, drawing on the same judgment the underwriter used to write the account. Analytics can warn you — a deteriorating work-in-process schedule, a slipping completion percentage, a spike in supplier complaints can flag a contractor sliding toward default earlier than an annual review would. That early warning is genuinely valuable and increasingly data-driven. But the warning only tells you to look; what to do remains a judgment exercised under pressure, on incomplete information, with millions at stake. Surety is the clearest case in the book of technology as the underwriter's instrument rather than the underwriter's replacement.

A final word on the combined ratio, to tie this line back to the number that judges every line (Chapter 3 owns it). Because surety underwrites toward zero expected loss and recovers much of what it does pay, a well-run surety operation can post a very low loss ratio and a strong combined ratio across a normal stretch of years — and that is exactly what makes the line attractive when it is written with discipline. But the losses, when they come, are lumpy and correlated with the economy: a recession that squeezes construction can push several contractors into default at once, and a single large contractor's failure can produce a loss that swamps a year's fees. The surety's combined ratio is therefore calm for long stretches and then violent — the opposite of a high-frequency line's steady churn. Reading that pattern correctly is the discipline of surety portfolio management: you do not chase the line's low average loss ratio into overcapacity, because the average conceals the tail, and the tail is where sureties die. Pricing follows risk imperfectly in surety (theme four, again, bent by the structure of the line); selection and capacity discipline are what keep the combined ratio honest, far more than the fee.

🗂️ The Underwriting File

Surety as an adjacency — what Harbor Steel would need if it bid public work. Harbor Steel is, in this book, primarily a property-casualty account: you have been building its property, general-liability, workers'-compensation, and commercial-auto coverages across this part, and surety is not part of the core program Meridian submitted. But the account is worth a surety aside, because it sharpens the line between insuring a risk and guaranteeing a performance — and because a structural-steel fabricator is exactly the kind of business that may one day need bonds.

Suppose Harbor Steel decides to bid a piece of public work — say, fabricating and erecting the structural steel for a municipal facility under a contract with a city or a state agency. The moment it does, the Little Miller Act in its state will almost certainly require it to furnish a bid bond with its proposal and, on award, a performance bond and a payment bond, each at or near the contract price. Those bonds are not insurance and would not come from the property-casualty side of your house; they would come from a surety, after a credit underwriting of Harbor Steel as a principal.

Now watch how differently the same company reads through the surety lens. On the property side, Harbor Steel's aging roof, its two fires, and its named-windstorm exposure are the story — the hazards you have spent twelve chapters analyzing. The surety underwriter barely cares about any of that. The surety underwriter wants Harbor Steel's financial statements: its working capital (can it fund the labor and steel before the city pays?), its net worth (can it absorb a bad job?), its backlog (is this public job on top of a fabrication shop already running near capacity?), and its bank line (does it have a liquidity backstop?). The surety wants to know whether Harbor Steel has erected structural steel on public projects of this size before — its capacity — and what its owners' character and payment record look like. And the surety will require a General Indemnity Agreement signed by the company and its single owner personally, putting the owner's own assets behind the guarantee.

What this aside settles, and what it does not. It settles a conceptual boundary: the bonds Harbor Steel would need to bid public work are credit instruments, underwritten toward zero loss on the strength of the balance sheet and the owner's character and indemnity — a different decision, by a different underwriter, on different evidence, from the property-casualty package you are assembling. It does not change the Harbor Steel disposition at all: surety remains outside the submitted program, a teaching adjacency rather than a coverage we are adding. If the owner later pursues public work, the account would acquire a surety relationship alongside its insurance — and the file would gain a credit dimension it does not have today. The running disposition is unchanged: this is a teaching aside on the surety adjacency, noted and set aside.

Conclusion

Surety looks like insurance and is sold by insurance companies, but it is credit — a three-party guarantee in which the surety stands behind a principal's promise to an obligee, and, having paid the obligee in the rare event of a default, reaches back to the principal for reimbursement. That one structural fact reorganizes everything: the loss ratio runs toward zero rather than toward a profitable 60-something percent; the "premium" is a fee for the use of credit, not the price of an expected loss; the file's center of gravity is a financial statement read like a banker's, not a loss run read like an underwriter's; and the worst case is not a covered peril but a business failure that triggers an expensive completion and a fight for salvage. The two great families — contract surety (bid, performance, payment bonds on construction) and commercial surety (license, court, and fiduciary bonds on a thousand other promises) — share that credit structure even as their bonds guarantee wildly different obligations.

The discipline that prices and selects these risks is lending discipline, compressed into the three C's: character, capacity, and capital, weighted in surety in roughly that order, because no balance sheet saves a job a dishonest or overreaching contractor is botching. We read the financials for working capital, net worth, backlog, and the bank line; we read the people and the track record for character and capacity; we secured the whole line with the General Indemnity Agreement that makes reimbursement possible; and we followed a defaulted bond through the surety's options and into salvage, to see why a surety loss behaves like a lender's loss and why the line's calm combined ratio hides a violent, economy-correlated tail. Above all, this chapter showed you a place where the book's central conviction is not an argument but a fact: surety is the line where judgment is genuinely irreplaceable, because its decisive variable — whether to trust a contractor on a job — is one no model can read.

In the next chapter we leave the credit world and return to insurance proper, opening the door to the specialty and niche lines — ocean and inland marine, aviation, energy, crop, and the program business — where each corner of the market has its own logic, its own hazards, and its own kind of underwriter. Harbor Steel will reappear there too: the steel it ships and the equipment it hauls are an inland-marine exposure, and the package we have been building has one more piece to add.

Key Terms

• Surety bond — a three-party contract in which a surety guarantees to an obligee that a principal will fulfill an obligation; if the principal fails, the surety performs or pays the obligee and then seeks reimbursement from the principal, making the instrument a guarantee (credit) rather than a transfer of risk.
• Three-party relationship — the structure of every surety bond, comprising the principal (who must perform the obligation), the obligee (who is protected and receives the obligation), and the surety (who guarantees the principal's performance to the obligee); the payer (principal) is not the protected party (obligee).
• Contract surety — bonds guaranteeing a contractor's obligations on a construction or service contract — principally the bid bond (the contractor will honor its bid and post the bonds), the performance bond (the work will be completed per the contract), and the payment bond (subcontractors and suppliers will be paid).
• Commercial surety — the broad category of non-contract bonds guaranteeing other obligations, grouped as license and permit bonds (regulatory compliance), court bonds (judicial and fiduciary), and public official and miscellaneous bonds.
• The surety "three C's" — the credit-underwriting framework for a surety account: character (the principal's integrity and track record), capacity (its ability to perform the work at the size and volume contemplated), and capital (its financial strength — working capital, net worth, liquidity, and access to credit).

Spaced Review

1. Explain why a surety underwrites toward a near-zero expected loss, and contrast that with how a property insurer prices an expected loss into its premium. What single structural fact about surety produces this difference? (§25.1)
2. A contractor's financial statement shows a healthy reported working capital, but a large share of its current assets is a receivable from another company the same owner controls. How would a surety adjust the working-capital figure, and why? (§25.5)
3. Recall from Chapter 1: define adverse selection in your own words, then explain how it shows up differently in surety — who is the dangerous applicant in a surety account, and why can't you price your way out of the danger? (Ch. 1; §25.4)
4. From Chapter 6, frequency × severity is the structure of expected loss. Why does this framing serve a surety underwriter so much worse than it serves an auto or property underwriter, and what does the surety underwriter rely on instead? (Ch. 6; §25.4, §25.7)
5. (The recurring pricing-discipline question.) A soft market is pushing your competitors to raise contractors' bonding programs beyond their demonstrated capacity, and you are losing accounts by holding the line on single-job and aggregate limits. Would chasing those programs with a higher bond fee help or hurt your combined ratio over the cycle, and why is "selection and capacity discipline," not price, the real protection in this line? (Ch. 3; §25.4, §25.7)