Signing beforehand increases honesty.

Here is a fascinating simple study by Shu et al. Simply signing a statement of honesty at the top rather the bottom of a form makes you more honest. They outline the problem:

The annual tax gap between actual and claimed taxes due in the United States amounts to roughly $345 billion. The Internal Revenue Service estimates more than half this amount is due to individuals misrepresenting their income and deductions (1). Insurance is another domain burdened by the staggering cost of individual dishonesty; the Coalition Against Insurance Fraud estimated that the overall magnitude of insurance fraud in the United States totaled $80 billion in 2006 (2). The problem with curbing dishonesty in behaviors such as filing tax returns, submitting insurance claims, claiming business expenses or reporting billable hours is that they primarily rely on self-monitoring in lieu of external policing.

Here is their abstract:

Many written forms required by businesses and governments rely on honest reporting. Proof of honest intent is typically provided through signature at the end of, e.g., tax returns or insurance policy forms. Still, people sometimes cheat to advance their financial self-interests—at great costs to society. We test an easy-to-implement method to discourage dishonesty: signing at the beginning rather than at the end of a self-report, thereby reversing the order of the current practice. Using laboratory and field experiments, we find that signing before, rather than after, the opportunity to cheat makes ethics salient when they are needed most and significantly reduces dishonesty.

The experimental design used several different measures of cheating: self-reported earnings (income) on a math puzzles task wherein university participants could cheat for financial gain, self reported travel expenses to the laboratory (deductions) claimed on a tax return form on research earnings. Another experiment was done in the field with an insurance company in the southeastern United States asking some of their existing customers to report their odometer reading. They examined the effect of requiring the signature at the top of the form, the bottom, or the control of requiring no signature.