Madman or TACO-Man: What we can learn from Trump

Threats are the loudest currency in negotiations, and at the same time the most fragile. Donald Trump’s approach to Iran is a prime example of how quickly a policy of threats can lose its effectiveness once it becomes predictable.

A list of the last six publicly documented episodes involving Iran illustrates this: harsh rhetoric, an ultimatum, dramatic escalation and, in the end, a reprieve, an offer of talks or, at most, limited action instead of the announced ‘final’ consequences.

1. On 22 March, he issued an ultimatum regarding the Strait of Hormuz, accompanied by an announcement that he would attack Iranian power stations. This was followed by deadline extensions – but no action was taken.
2. On 7 April, the rhetoric became even more drastic (“wipe out an entire civilisation”), but shortly afterwards a ceasefire was declared and the markets calmed down.
3. On 20 and 21 April, too, statements such as “I expect to bomb” were not followed by an attack, but by an extension of the ceasefire “indefinitely”.
4. On 29 April, Trump staged his threat with “No more Mr Nice Guy” and an AI image; militarily, nothing happened.
5. On 4 May, he spoke in a TV interview of Iran being “wiped off the face of the earth”; shortly afterwards, “Project Freedom” ended with reference to diplomacy.
6. And even after the announcement on 7 May that he would now proceed “much harder and much more violently”, defensive counter-strikes followed, but no large-scale attack.

In the US, an acronym has long been circulating for precisely this pattern: TACO – “Trump Always Chickens Out”, meaning: threaten as much as possible, then back down.

A recurring pattern

Every threat contains a strategic crux: it only works as long as it remains unclear whether it will actually be carried out. In negotiations, what counts is not how tough something sounds – but how credible the costs are that will actually materialise if the other side does not give in. Credibility arises from consistency, capability and a willingness to act.

But if, over several rounds, it is observed that:

• deadlines are missed,
• ultimatums are renegotiated,
• de-escalation is regularly the final outcome,

…then something crucial happens: the other side dismisses the threat as ‘noise’. And they can even play strategically with it: wait and see, let the threat-maker ‘fly off the handle’, factor in the next extension.

That is the paradoxical twist of threats: they are meant to demonstrate the ability to act and power. The risk, however, is that if they are not carried out, they prove the opposite. Then every new threat becomes more costly: one has to raise one’s voice even louder just to have any effect at all. This drives escalation spirals upwards and credibility downwards. Predictability becomes the consequence. Yet unpredictability can also be an instrument of power.

The Madman Theory: Unpredictability as an Instrument of Power

The Madman Theory is often attributed to former US President Richard Nixon. The basic idea is simple: an opponent should believe that the person making the threat may be acting irrationally and therefore will not shy away from extreme measures.

Those perceived as unpredictable do not need to constantly prove their threats. The mere possibility of escalation is enough to make the opponent more cautious.

Trump is a great proponent of this theory and has cultivated this principle over the years. In his ‘America First’ speech, he said: ‘We must as a nation be more unpredictable’. In 2024, Trump also articulated this logic very directly with regard to China and Taiwan: he said that military force against a Chinese blockade of Taiwan was unnecessary because Xi Jinping respected him and knew that he was “f— crazy”. His political opponents, allies and negotiating partners should never know exactly what would happen next. The uncertainty itself became a strategic tool.

Yet this is precisely where the strategy’s weakness lies: unpredictability only works as long as it remains credible.

If it is repeatedly observed that drastic announcements are not followed by corresponding actions, perceptions change. The supposedly irrational actor becomes a predictable one. The opponent recognises a pattern. Instead of fear of an uncontrollable escalation, the expectation
arises that de-escalation will ultimately occur once again. The ‘madman’ becomes an actor whose script is transparent.

Threats are not just ‘tough’; they are highly toxic negotiating tools

Anyone who escalates to the maximum at the negotiating table increases the risk of miscalculations and cuts off their own retreat. At the same time, the conflict shifts away from interests towards identity and prestige. Giving in then feels like humiliation; compromises become politically and psychologically more costly. And every threat that goes unfulfilled erodes one’s own credibility. Eventually, the only choice left is between an even more drastic tone or a visible loss of authority. There is also another effect: those who make pressure the standard language teach their counterpart to think only in terms of pressure as well.

Threats are not simply the harsh version of argumentation. They are a risky tool that can quickly do more harm than good in the short term.

Threats are not a standard tool in negotiations, but a high-risk instrument. Those who use them regularly turn them into noise. And noise can be ignored.

Learning from Trump means learning to fail.

Negotiations are learning systems. The other side observes not only what is said, but also the consequences that follow. If maximal rhetoric is repeatedly followed by deadline extensions, offers of dialogue or limited, symbolic responses, a pattern emerges. This pattern can be analysed, planned and exploited. The threat is then no longer assessed as a serious cost risk, but as part of a choreography that reliably culminates in an evasive manoeuvre. Pressure becomes expectation management.

This has three consequences. Firstly, the reaction shifts from respect to tactics: one waits and sees, tests boundaries, drags things out and factors in the next backtracking. Secondly, every new threat must be more drastic in order to attract any attention at all. As a result, the rhetoric of escalation rises faster than the actual willingness to escalate, making miscalculations more likely. Thirdly, one’s own position erodes structurally because credibility is lost not gradually but in leaps and bounds as soon as words and actions diverge.

The key lesson is therefore clear and simple. Those who make threats a routine provide their opponents with data points. Those who provide data points become predictable. And those who are predictable lose the only resource that could ever sustain threats: the doubt as to whether action will actually be taken this time. This is precisely how demonstrated toughness turns into a calculable weakness. Learning from Trump means learning to fail, because in the process one learns how to burn through credibility.