Several Modi-era accountability concerns are real, but viral reel overstates causation and superlatives

A 7 June 2026 Instagram Reel by advocate Harshvardhan Singh Thakur presents a rapid list of criticisms of Prime Minister Narendra Modi under the label “Fact Check”. The clip combines claims about press conferences, comedians and detention, examination leaks, fuel prices, the rupee, inflation, women’s safety, unemployment and hunger, repeatedly describing Modi as the country’s “first” prime minister to produce or permit these outcomes.

The strongest finding is not one blanket verdict on the Reel. Several underlying concerns are real and independently measurable, but they belong to different institutions, time periods and causal chains. The absence of a conventional prime-ministerial press conference taking open questions is well supported; the rupee reached successive nominal record lows in 2026; fuel prices were materially above 2014 levels; Munawar Faruqui was jailed in a state-police case; and Gurmeet Ram Rahim Singh was on parole when the Reel was posted. At the same time, the Reel turns political estimates into settled statistics, uses undefined “first” and “peak” superlatives, and sometimes attributes outcomes directly to the prime minister without evidence establishing that causal link.

The public-interest issue is accountability with correct institutional attribution

The Reel’s subjects are politically connected because they concern government performance and democratic accountability, but evidentially they are not interchangeable. A prime minister’s media-access practice is directly attributable to the prime minister. A state-police arrest may occur during the same national government without being ordered by it. Public-examination leaks can involve Union or state agencies and private networks. Fuel prices reflect taxes, crude prices, exchange rates and policy choices. The rupee responds to oil prices, capital flows, interest-rate expectations and Reserve Bank intervention. Unemployment and hunger depend on the selected measure and period.

Collapsing these into a single personal scorecard creates two opposite risks: genuine accountability failures can be dismissed because one statistic is weak, or weak claims can acquire credibility from unrelated facts that are true. A defensible record therefore has to separate conduct, institutional responsibility, trend and causation before drawing a wider conclusion.

Press conferences: the factual core is strong

The Reel says Modi did not hold a press conference during his tenure. That is substantially supported if “press conference” means a conventional, open event in which the prime minister takes unscripted questions from journalists. Modi has given many interviews, speeches and media appearances, and he attended a BJP press conference in 2019 where Amit Shah handled the questions, but reporting through 2026 continued to describe him as not holding a traditional press conference taking journalists’ questions as prime minister.

This is not merely an opposition formulation. In June 2026 BJP MP Tejasvi Surya defended the practice by arguing that social media and direct communication had made press conferences redundant. In August, Parliamentary Affairs Minister Kiren Rijiju similarly defended Modi’s preference for direct outreach and interviews. The underlying democratic question is therefore not whether the prime minister communicates at all, but whether direct communication is an adequate substitute for a forum in which independent journalists can choose and pursue questions. That is a normative question about scrutiny and accountability, not proof of concealment or improper motive.

Paper leaks: a real governance problem, but “90+” is not an audited national count

The Reel states that more than 90 examination papers leaked during Modi’s tenure. The number can be traced to political claims, including a 2026 Congress allegation of roughly 90 leaks over about 12 years, but IndiaFiles has not identified an auditable national list that defines which examinations, jurisdictions and incidents are counted and how repeat or related cases are handled.

The broader issue is nevertheless real. Parliament enacted the Public Examinations (Prevention of Unfair Means) Act in 2024, covering major Union recruitment and entrance examinations. In July 2026 the government introduced amendments proposing stronger penalties, special investigative arrangements and faster trials, while explicitly acknowledging continuing paper-leak and examination-malpractice cases. The existence of legislation and later strengthening measures supports a serious exam-integrity problem; it does not validate the Reel’s exact “90+” total or make the prime minister personally responsible for every state and Union case.

Fuel and the rupee: the trend is real; one-person causation is not

Official petroleum data place Delhi petrol at ₹71.41 per litre in April 2014 and diesel at ₹56.71 in May 2014. In May 2026, amid a West Asia oil shock, retail fuel prices rose again and Delhi petrol moved above ₹100 per litre. The Reel’s “touching the sky” language is rhetorical, but the direction of the long-run nominal comparison is factual.

The rupee likewise depreciated sharply in nominal terms. Government data put it at ₹58.59 per US dollar on 26 May 2014; by early June 2026 it was near ₹95 after a sequence of record lows. But exchange rates are not a single-policy score. Oil prices, foreign investment flows, the US dollar, interest-rate expectations, inflation differentials and Reserve Bank action all matter. The evidence supports the statement that the rupee reached historic nominal lows during Modi’s tenure; it does not by itself establish that the prime minister caused the depreciation.

Jail and parole comparisons need date and authority

The Reel says comedians were jailed. Munawar Faruqui provides a documented example: he spent more than a month in jail after a 2021 arrest by Madhya Pradesh police before the Supreme Court granted interim bail. That makes the blanket dismissal of the comedian claim wrong, but direct attribution to Modi is too broad because the arrest was a state-police action and the Reel does not establish a prime-ministerial order or policy causing it.

The Reel also contrasts Sonam Wangchuk with convicted Dera Sacha Sauda chief Gurmeet Ram Rahim Singh. This comparison was already stale when the Reel was posted. Wangchuk, detained under the National Security Act in September 2025, was released after the Union government revoked his detention on 14 March 2026. Ram Rahim, serving a 20-year sentence for raping two disciples, was released on a 30-day parole on 26 May 2026, making this his 16th temporary release since 2020. On 7 June, therefore, Ram Rahim was outside prison but Wangchuk was not in jail. The contrast can raise legitimate questions about detention and temporary-release policy, but the Reel’s present-tense formulation was inaccurate and the two legal processes are not equivalent.

Unemployment and hunger: “at their peak” is not established

The Reel says unemployment and hunger were at their peak. Neither claim is usable without a defined metric, population and comparison period. The 2025 Periodic Labour Force Survey, published in March 2026, reported broadly stable labour-force participation and worker-population ratios and a fall in usual-status unemployment among people aged 15 and above with secondary education or more, from 7.0% in 2024 to 6.5% in 2025. This does not mean labour-market distress is absent—youth and educated unemployment can be much higher—but it contradicts an undefined claim that unemployment as a whole was at an all-time peak.

Hunger is similarly multidimensional. The Global Hunger Index combines undernourishment and child-nutrition and mortality indicators and cautions against treating annual ranks from changing editions as a simple time series; the Indian government disputes aspects of its methodology. India continues to face serious nutrition problems, but the Reel supplies no measure capable of establishing that “hunger” was at its historical peak.

Inflation and women’s safety are concerns, not verifiable absolutes in the form used

“Inflation broke the public’s back” is a political judgment about household hardship, not a defined statistical claim. A meaningful test would require a selected CPI period, food and fuel components, wage or income movement and distribution across households. Likewise, saying the prime minister “never answered” on women’s safety is an absolute communication claim that would require a defined universe of speeches, interviews or parliamentary responsibility. Crime against women and women’s safety are major public-policy issues, but those facts do not prove the Reel’s absolute wording.

The unidentified “rapist in government” allegation remains unresolved

The Reel refers to a rapist being kept “in his government” but does not identify the person. IndiaFiles cannot safely infer the target from the surrounding rhetoric. Ram Rahim is separately named later in the clip and was not a member of the Modi government. Without an identifiable office-holder and evidence of both the relevant conviction and government membership, this allegation remains unverified as stated.

What the Reel gets right—and what it cannot prove

The Reel works as political rhetoric because it places many grievances next to one another, but proximity is not evidence of a common mechanism. The conventional-press-conference criticism is strongly supported. Paper leaks are a recognised institutional problem even though the “90+” number is not independently established. Fuel prices were materially higher than in 2014 and the rupee reached nominal record lows. Faruqui’s imprisonment is real, while the Wangchuk/Ram Rahim contrast was temporally wrong on the Reel’s publication date. Claims about “first prime minister”, peak unemployment, peak hunger, never answering on women’s safety and an unidentified rapist in government remain unproven or too imprecise.

The wider significance is methodological but consequential: political accountability becomes more reliable when responsibility is assigned to the institution and decision actually evidenced. A true economic trend does not prove personal causation; a real arrest does not prove a prime-ministerial order; a recurring exam-security problem does not validate an unaudited total. Conversely, exaggeration in one part of a viral political argument should not be used to erase the well-supported concerns in another. The record supports a mixed assessment, not a single verdict on the government or the Reel.