Imposing sanctions on non-compliant parties to international agreements is advocated as a remedy for international cooperation failure. Nevertheless, sanctions are costly, and rational choice theory predicts their ineffectiveness in improving cooperation. We test sanctions effectiveness experimentally in international collective-risk social dilemmas simulating efforts to avoid catastrophic climate change. We involve individuals from countries where sanctions were shown to be effective (Germany) or ineffective (Russia) in increasing cooperation. Here we show that, while this result still holds nationally, international interaction backed by sanctions is beneficial. Cooperation by low cooperator groups increases relative to national cooperation and converges to the levels of high cooperators. This result holds regardless of revealing other group members' nationality, suggesting that participants' specific attitudes or stereotypes over the other country were irrelevant. Groups interacting under sanctions contribute more to catastrophe prevention than what would maximise expected group payoffs. This behaviour signals a strong propensity for protection against collective risks.

The dataset was collected in lab experiments conducted with the Ztree software (Fischbacher, U. 2007 z-Tree: Zurich toolbox for ready-made economic experiments. Exp. Econ. 10, 171-178.) Raw data have been assembled in csv format and analysed with Stata17.

Analyses have been conducted with Stata 17. A codebook for the variables included in the dataset is uploaded.

Raw data are provided in csv format. Missing observations are identified with a dot, namely with a . sign.

Codes to reproduce econometrics results, computation of effect sizes and figures

can be found in the file "Analyses codes for reproduction.do", 

of which a text version has also been uploaded.

A pdf version of the stata output has been uploaded.

The file Codebook.xlsx includes a description/definition of the variables 

included in the dataset.  

The codes in the file "Analyses codes for reproduction.do" are organized as follows:

- From line 106: REGRESSIONS USED IN TABLE 2 OF MAIN PAPER AND TABLE S11 IN 

  Supplementary Informatio (SI) (Analysis of impact of sanctions - expressed as 

  number of tokens deducted -on cooperation)

- From line 244: REGRESSIONS REPORTED IN TABLE S12 IN SM

 (Analysis of impact of sanctions - expressed as dichotomous variable 

  identifying whether somone has been sanctioned or not - on cooperation)

- From line 325: SANCTION AS DEP VAR: TABLE S14

  (Analysis of determinants sanctions)

- From line 367: SANCTION AS DEP VAR: TABLE S13

  (Analysis of determinants sanctions)

- From line 539: EFFECT SIZES AND TESTS

- From line 1029: FIGURES 2 and 3

The dataset is to be intended as property of the authors and 

the only allowed use is for reproduction of the results included in the paper.

No other use is allowed.