📈 economics

Demand-Mediated Rather Than Productivity-Led Wages: Evidence from Albania

Using Albanian data from 2000–2024, this paper demonstrates that annual wage dynamics are more consistently driven by demand-side factors like GDP, consumption, and remittances than by a direct, stable pass-through of measured labour productivity, suggesting that in small, open economies, wage growth is primarily demand-mediated rather than productivity-led.

Original authors: Admir Mulaj, Adela Karapici

Published 2026-07-15

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Admir Mulaj, Adela Karapici

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the Albanian economy as a bustling, growing city where everyone is trying to figure out why workers' paychecks are getting bigger. For a long time, the standard rule of the game was simple: Productivity First. The idea was that workers get a raise only after they prove they can build more, sell more, or work faster. It's like a video game where you only unlock the next level of rewards after you've beaten the boss.

But this paper, written by Admir Mulaj and Adela Karapici, looks at the data from Albania between the years 2000 and 2024 and says, "Wait a minute. The game isn't working that way here."

The Main Discovery: The "Party" vs. The "Workout"

Instead of paychecks rising because workers suddenly got super-strong (productivity), the paper suggests wages in Albania are rising because the whole economy is having a party (demand).

Think of the economy as a giant house party.

The Productivity Theory says: "You only get more food (wages) if you help cook more meals (productivity)."
The Paper's Finding says: "Actually, you're getting more food because the fridge is full, the guests are spending money, and the host has a big budget (GDP, remittances, and investment)."

The researchers found that Albanian wages move in lockstep with the size of the party—how much money is flowing in, how much people are spending, and how much the government is building—not necessarily with how much each individual worker is producing in a given year.

What the Paper Rules Out

The authors are very careful to say what this study is not saying.

It is NOT a proof that productivity doesn't matter. They explicitly state that productivity is still the engine for long-term success. If you want to stay rich forever, you need to build better machines and skills.
It is NOT a claim that GDP forces wages up like a robot. They reject the idea that a 1% jump in the country's total money (GDP) mechanically creates a specific, guaranteed jump in wages.
It is NOT a magic bullet. The paper argues against the idea that we can just look at a worker's output and instantly know their pay. In Albania's annual data, that link is blurry and hard to spot.

The "Fuzzy" Connection: How Sure Are They?

The authors are like detectives who found a very strong clue, but they admit the crime scene is a bit messy. They are suggesting a pattern, not proving a law of physics.

Here is the evidence they gathered:

The "Party" Variables: When they looked at things like Gross Domestic Product (GDP), final consumption (people buying stuff), investment, and remittances (money sent home by Albanians working abroad), these numbers danced right along with wages.
The "Workout" Variable: When they tried to isolate labor productivity (how much one person produces), it didn't show up as a clear, steady driver of wages in their yearly data. It was like trying to hear a single violin in a rock band; the sound of the whole band (the economy) was too loud.
The "Public Sector" Lag: They found that government workers' pay often reacts with a delay. It's like the government waits for the budget to be approved or for inflation to hit the headlines before giving raises. This happens about two to three years after the economic boom starts.
The "Migration" Factor: The paper notes that because many Albanians can move to Europe for work, they have an "outside option." If the local party isn't paying well, they might leave. This pressure helps keep wages up, but the data on migration is too short and messy to be the main star of the show.

The Numbers and the Timeline

The study uses annual data from 2000 to 2024.

They found that in some models, the connection between GDP and wages was so strong that a 1% increase in GDP seemed linked to a larger than 1% increase in wages. But the authors warn us: don't take that literally as a rule. It's more like a signal that when the whole economy expands, wages tend to follow, but the math is a bit "reduced-form" (a fancy way of saying it's a summary of many things happening at once, not a direct cause-and-effect switch).
They checked their work using international data from the World Bank (specifically the World Development Indicators) to make sure they weren't just making things up with their own numbers. The story stayed the same: demand and spending matter more than the isolated productivity number.

The "European Neighbor" Effect

The paper also points out that Albanian workers aren't just looking at their own boss; they are looking at their neighbors in Germany, Italy, and Croatia.

Imagine a worker in Albania comparing their pay to a cousin in Germany. Even if they can't move, that comparison changes what they think is a "fair" wage.
The paper notes that the ratio of Albania's minimum wage to Germany's minimum wage jumped from about 0.11 in 2015-2016 to about 0.19 in 2024. This gap closing suggests that the pressure to match European standards is real, even if the data is hard to measure perfectly.

The Bottom Line

The paper concludes that Albania is a "demand-mediated" economy. This means wages are driven by the flow of money, the size of the party, and the pressure to keep workers from leaving, rather than a strict, immediate rule that says "more work = more pay."

The authors suggest that while this has helped Albanians earn more and live better, it's a bit fragile. If the party stops (if spending slows down or remittances dry up), wages might drop unless the country also builds better "machines" (productivity) and stronger rules for how pay is decided (institutions). They aren't saying the current system is broken, just that it works differently than the old textbooks predicted. It's a story of a country catching up, where the crowd's energy drives the paychecks, not just the individual's speed.

Technical Summary: Demand-Mediated Rather Than Productivity-Led Wages: Evidence from Albania

Problem Statement
This study investigates the determinants of wage formation in Albania, specifically addressing whether annual wage dynamics are better characterized by a "clean" productivity-pass-through mechanism or by a broader, demand-mediated process. Since 2013, Albania has experienced substantial real-wage growth alongside limited measured labor-productivity convergence. This divergence challenges the standard marginal-productivity theory in the context of a small, open, migrant-sending economy. The central empirical question is not whether productivity matters in isolation, but whether Albanian wage dynamics are more consistent with a demand-mediated pattern—shaped by aggregate demand, fiscal capacity, remittances, and migration—than with a stable, productivity-led anchor in annual macro data.

Methodology
The analysis utilizes annual data spanning 2000–2024. Wage variables are derived from author-compiled series based on INSTAT (Albanian Institute of Statistics) data, while macroeconomic controls are sourced from the World Bank's World Development Indicators (WDI). The study employs the Autoregressive Distributed Lag (ARDL) bounds-testing approach and Unrestricted Error-Correction Model (UECM) specifications to estimate dynamic relationships.

The empirical strategy is organized around a strict hierarchy of evidence to manage the limitations of a short time series (effective $N \approx 21$ after lag adjustment) and high multicollinearity among macro variables:

GDP-Centered Models: Primary specifications using real GDP, final consumption, investment, and remittances as proxies for domestic absorption and macroeconomic capacity.
Labor-Productivity-Centered Models: Direct comparators testing the stability of productivity as an independent wage anchor.
Combined Specifications: Stress tests including both GDP and productivity simultaneously to assess identification under severe multicollinearity.
Robustness Checks: Additional tests using standardized WDI series, lag-depth screening (0–5 years), and Eurostat wage-reference benchmarks to contextualize migration and outside options.

The study treats GDP not as a mechanical structural cause but as a reduced-form proxy for the macroeconomic environment (domestic absorption, fiscal space, construction/tourism cycles). Diagnostic checks include HAC (Newey-West) standard errors, bounds tests for cointegration, and variance inflation factor (VIF) analysis to interpret coefficient instability.

Key Results

Demand-Mediated Association: The results indicate that Albanian wages are more consistently associated with GDP, final consumption, investment, and remittances than with measured labor productivity as a separately stable annual anchor. In GDP-centered specifications, private wages show a strong long-run association with the macro-demand block, with implied GDP elasticities often exceeding unity (interpreted as reduced-form indicators of a common catching-up trend rather than structural elasticities).
Instability of Productivity Coefficients: In productivity-centered and combined models, labor productivity coefficients are frequently small, unstable, insignificant, or negatively signed. This suggests that in annual aggregate data, productivity does not provide a clean, independent wage anchor once investment, prices, and wage inertia are controlled for. The instability is attributed to multicollinearity and the difficulty of separating productivity from the broader output trend in a catching-up economy.
Public vs. Private Dynamics: Public wages respond to macro-fiscal conditions but exhibit delayed adjustment (e.g., reacting to GDP growth at a two-year lag), reflecting budget procedures and political cycles. The public-private wage differential is driven more by inflation and unemployment (labor market slack) than by output or productivity levels.
Migration and External Benchmarks: While migration data is too short for full structural identification, correlation analysis and Eurostat benchmark tests suggest that European wage levels act as a reference frontier. This influences reservation wages and retention pressures, creating a demand-mediated channel where wage claims adjust to external benchmarks even when domestic productivity signals are weak.
Lag Structure: Average real wages move contemporaneously with demand and absorption variables (lag 0). In contrast, public wages and capacity indicators (education, technology) show evidence of delayed adjustment (lags 1–3), supporting a view of institutional mediation rather than immediate market clearing.

Significance and Claims
The paper claims to contribute to European debates on wage-productivity decoupling by documenting a disciplined empirical regularity in a small, open, transition economy: wages align more closely with a broad demand block than with measured productivity alone.

Theoretical Positioning: The findings support a Post-Keynesian and Kaldorian reading of wage formation, where effective demand and institutions condition wage growth more visibly than measured productivity in the short-to-medium run. The study argues that Albania operates in a demand-constrained environment shaped by weak collective bargaining, outward migration, and public-sector wage signaling.
Policy Implications: The paper posits that demand-supported wage growth (driven by remittances, construction, tourism, and public spending) has improved living standards but remains fragile without accompanying productivity-enhancing investments and institutional reforms. The central policy challenge is not a choice between demand and productivity, but the construction of institutions (formalization, collective bargaining, transparent public wage rules) that can translate demand-supported gains into durable wage-productivity sharing.
Limitations and Modesty: The authors explicitly caution against interpreting the results as structural causal proof. They emphasize that the high fit of GDP-centered models and the instability of productivity coefficients reflect the difficulty of identifying separate channels in a highly collinear, trending annual dataset. The study concludes that while productivity is essential for long-run convergence, its transmission into wages in Albania is institutionally filtered and difficult to isolate in aggregate annual data. The evidence supports a "demand-mediated" interpretation rather than a "GDP-only" or "productivity-led" rule.